drone_detection_YOLOv7x.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Drone Detection Using YOLOv7: A Comprehensive Dataset and Model Approach\n",
        "\n",
        "<br>\n",
        "\n",
        "<img src=\"https://images.pexels.com/photos/442587/pexels-photo-442587.jpeg\" height=550 width=1000 alt=\"https://www.pexels.com/\"/>\n",
        "\n",
        "<small>Picture Source: <a href=\"https://images.pexels.com/photos/\">pexels</a></small>\n",
        "\n",
        "<br>\n",
        "\n",
        "## Abstract\n",
        "Unmanned aerial vehicles (UAVs), commonly known as drones, have become increasingly prevalent in various domains, including surveillance, photography, and delivery services. However, the rapid proliferation of drones raises concerns regarding security and privacy threats. To address these concerns, effective drone detection systems are crucial for identifying and tracking drones in real-time. In this research, we present a comprehensive dataset and propose a state-of-the-art drone detection model using the YOLOv7 architecture.\n",
        "\n",
        "## Introduction\n",
        "The widespread adoption of drones has led to an urgent need for reliable drone detection systems to ensure the safety and security of public spaces. Drone detection poses unique challenges due to the small size, fast movement, and diverse appearance of drones, making traditional object detection methods insufficient. Therefore, there is a growing demand for advanced detection models that can accurately identify drones in complex environments.\n",
        "YOLO (You Only Look Once) is an object detection algorithm that was introduced in 2015 by **Joseph Redmon** et al. It revolutionized the field of computer vision by providing a real-time object detection solution with impressive accuracy.\n",
        "\n",
        "\n",
        "## Dataset\n",
        "To facilitate the development and evaluation of drone detection models, we introduce a novel and comprehensive dataset specifically curated for training and testing drone detection algorithms. The dataset, sourced from the publicly available [\"YOLO Drone Detection Dataset\"](https://www.kaggle.com/datasets/muki2003/yolo-drone-detection-dataset) on Kaggle, comprises a diverse set of annotated images captured in various environmental conditions and camera perspectives. The dataset includes instances of drones along with other common objects to enable robust detection and classification.\n",
        "\n",
        "## Methodology\n",
        "In this study, we employ the YOLOv7 architecture, a popular and highly efficient object detection framework, for drone detection. YOLOv7 stands for \"You Only Look Once\" version 7, which utilizes a single neural network to simultaneously predict bounding boxes and class probabilities for multiple objects in an image. This architecture offers real-time performance, making it ideal for drone detection applications.\n",
        "\n",
        "## Experimental Setup\n",
        "To train and evaluate our drone detection model, we utilize the Colab platform, a cloud-based environment that provides access to powerful computing resources and deep learning libraries. Leveraging Colab's GPU acceleration capabilities, we train the YOLOv7 model using our curated dataset and fine-tune its parameters to optimize detection accuracy and efficiency.\n",
        "\n",
        "## YOLO\n",
        "\n",
        "\n",
        "*   **Single Pass Detection**: YOLO takes a different approach compared to traditional object detection methods that use region proposal techniques. Instead of dividing the image into regions and examining each region separately, YOLO performs detection in a single pass. It divides the input image into a grid and predicts bounding boxes and class probabilities for each grid cell.\n",
        "\n",
        "\n",
        "*   **Grid-based Prediction**: YOLO divides the input image into a fixed-size grid, typically, say, 7x7 or 13x13. Each grid cell is responsible for predicting objects that fall within it. For each grid cell, YOLO predicts multiple bounding boxes (each associated with a confidence score) and class probabilities.\n",
        "\n",
        "*   **Anchor Boxes**: To handle objects of different sizes and aspect ratios, YOLO uses anchor boxes. These anchor boxes are pre-defined boxes of different shapes and sizes. Each anchor box is associated with a specific grid cell. The network predicts offsets and dimensions for anchor boxes relative to the grid cell, along with the confidence scores and class probabilities.\n",
        "\n",
        "*   **Training**: YOLO is trained using a combination of labeled bounding box annotations and classification labels. The training process involves optimizing the network to minimize the localization loss (related to the accuracy of bounding box predictions) and the classification loss (related to the accuracy of class predictions).\n",
        "\n",
        "*   **Speed and Accuracy Trade-off**: YOLO achieves real-time object detection by sacrificing some localization accuracy compared to slower methods like Faster R-CNN. However, it still achieves competitive accuracy while providing significantly faster inference speeds, making it well-suited for real-time applications.\n",
        "\n",
        "<br>\n",
        "\n",
        "Since its introduction, YOLO has undergone several improvements and variations. Different versions such as YOLOv2, YOLOv3, and YOLOv4 have been developed, each introducing enhancements in terms of accuracy, speed, and additional features.\n",
        "\n",
        "It's important to note that this is a high-level overview of YOLO, and the algorithm has many technical details and variations. For a more in-depth understanding, it's recommended to refer to the original YOLO papers and related resources.\n",
        "\n",
        "\n",
        "## Keywords\n",
        "\n",
        "*   Drone detection\n",
        "*   YOLOv7\n",
        "*   Object detection\n",
        "*   Deep learning\n",
        "*   Surveillance\n",
        "*   Security\n",
        "\n",
        "<br>\n",
        "\n",
        "Make sure your runtime is **GPU** (_not_ CPU or TPU). And if it is an option, make sure you are using _Python 3_. You can select these settings by going to `Runtime -> Change runtime type -> Select the above mentioned settings and then press SAVE`.\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "q_LtQGXNQA6t"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Importing Libraries"
      ],
      "metadata": {
        "id": "Nut7NattQ-C6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import random\n",
        "import shutil\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "metadata": {
        "id": "qFRLkWriPzV0"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data Preprocessing"
      ],
      "metadata": {
        "id": "1z5v8sjlSrh0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rwMyw7FRY8v4",
        "outputId": "6437df1f-0607-43df-8cb7-023e15c62683"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip -q /content/archive.zip"
      ],
      "metadata": {
        "id": "5vbraRsGP7_b"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# !rm -rf /content/archive.zip"
      ],
      "metadata": {
        "id": "jpbeiWc3VTxW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@markdown ---\n",
        "#@markdown ### Enter image paths:\n",
        "train_images_dir = \"/content/drone_dataset/train/images\" #@param {type:\"string\"}\n",
        "val_images_dir = \"/content/drone_dataset/valid/images\" #@param {type:\"string\"}\n",
        "\n",
        "train_image_count = len([f for f in os.listdir(train_images_dir) if f.endswith(\".jpg\")])\n",
        "val_image_count = len([f for f in os.listdir(val_images_dir) if f.endswith(\".jpg\")])\n",
        "\n",
        "print(f\"Number of images in train folder: {train_image_count}\")\n",
        "print(f\"Number of images in val folder: {val_image_count}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tRpMs_fQW59X",
        "outputId": "835c76d8-256b-4aa3-f26c-5e06088880e1"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of images in train folder: 1012\n",
            "Number of images in val folder: 347\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@markdown ---\n",
        "#@markdown ### Enter label paths:\n",
        "train_labels_dir = \"/content/drone_dataset/train/labels\" #@param {type:\"string\"}\n",
        "val_labels_dir = \"/content/drone_dataset/valid/labels\" #@param {type:\"string\"}\n",
        "\n",
        "train_txt_count = len([f for f in os.listdir(train_labels_dir) if f.endswith(\".txt\")])\n",
        "val_txt_count = len([f for f in os.listdir(val_labels_dir) if f.endswith(\".txt\")])\n",
        "\n",
        "print(f\"Number of TXT files in train labels folder: {train_txt_count}\")\n",
        "print(f\"Number of TXT files in val labels folder: {val_txt_count}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r_KL7oxtZ-xv",
        "outputId": "ecea5ad2-9e95-4b77-f1a8-9e8646937bc8"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of TXT files in train labels folder: 1012\n",
            "Number of TXT files in val labels folder: 348\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Clone YOLO v7 and Train the Model"
      ],
      "metadata": {
        "id": "_OsRjWzAiMK4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xyeY9H6ccVMQ",
        "outputId": "0ea314fc-ad1e-436c-8128-d9cde5769b44"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Following URL of the GitHub repository that contains the YOLOv7 implementation by WongKinYiu. The .git extension indicates that it is a Git repository.\n",
        "\n",
        "When this command is executed, it will create a local copy of the entire repository on your machine. This allows you to access and use the YOLOv7 implementation provided by WongKinYiu for your own projects. You can then navigate to the cloned repository and access the relevant files and directories, including the training script, configuration files, and pre-trained weights."
      ],
      "metadata": {
        "id": "kcwa9kHqujYs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/WongKinYiu/yolov7.git"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S6bdPZ3JRaQN",
        "outputId": "feed0f4a-826e-45eb-ae8b-336cfa5290d3"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'yolov7'...\n",
            "remote: Enumerating objects: 1169, done.\u001b[K\n",
            "remote: Counting objects: 100% (30/30), done.\u001b[K\n",
            "remote: Compressing objects: 100% (21/21), done.\u001b[K\n",
            "remote: Total 1169 (delta 16), reused 18 (delta 9), pack-reused 1139\u001b[K\n",
            "Receiving objects: 100% (1169/1169), 70.42 MiB | 22.31 MiB/s, done.\n",
            "Resolving deltas: 100% (504/504), done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/yolov7"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kiTXUokERr-Q",
        "outputId": "e2ae54ed-2ed2-4af0-ac59-637a7527aebc"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov7\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can change model from [WongKinYiu GitHub](https://github.com/WongKinYiu/yolov7) page. When this command is executed, it downloads the \"yolov7x.pt\" file from the specified URL. This pre-trained weights file can then be used as a starting point for training or as a pre-trained model for inference in YOLOv7-based projects.\n",
        "\n",
        "Regarding other \".pt\" files for YOLO models, there are different variations of the YOLO architecture, each with its own pre-trained weights. Some commonly used pre-trained weights files for YOLO models include:\n",
        "\n",
        "* `yolov3.pt`: Pre-trained weights for YOLOv3.\n",
        "* `yolov4.pt`: Pre-trained weights for YOLOv4.\n",
        "* `yolov5s.pt`, yolov5m.pt, yolov5l.pt, yolov5x.pt: Pre-trained weights for \n",
        "\n",
        "YOLOv5, which is a different version developed by Ultralytics.\n",
        "These pre-trained weights files can be obtained from various sources, including official repositories, GitHub releases, or community contributions. It's important to note that the architecture and compatibility of the weights file should match the specific YOLO version you are using in your project to ensure proper functionality."
      ],
      "metadata": {
        "id": "SlmH3y-vSHF-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7x.pt"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dl2K_DqaRwnB",
        "outputId": "8a833731-6385-46c7-9b99-043c18b0090a"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-05-28 14:45:11--  https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7x.pt\n",
            "Resolving github.com (github.com)... 140.82.121.4\n",
            "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://objects.githubusercontent.com/github-production-release-asset-2e65be/511187726/c0e9f375-a42b-45d5-9e96-3156476cf121?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230528%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230528T144511Z&X-Amz-Expires=300&X-Amz-Signature=726fddd991016b82b0e32cdacf32f9f3fc3747402e5de90dd2cf28f4c7ca0b49&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=511187726&response-content-disposition=attachment%3B%20filename%3Dyolov7x.pt&response-content-type=application%2Foctet-stream [following]\n",
            "--2023-05-28 14:45:11--  https://objects.githubusercontent.com/github-production-release-asset-2e65be/511187726/c0e9f375-a42b-45d5-9e96-3156476cf121?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230528%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230528T144511Z&X-Amz-Expires=300&X-Amz-Signature=726fddd991016b82b0e32cdacf32f9f3fc3747402e5de90dd2cf28f4c7ca0b49&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=511187726&response-content-disposition=attachment%3B%20filename%3Dyolov7x.pt&response-content-type=application%2Foctet-stream\n",
            "Resolving objects.githubusercontent.com (objects.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.111.133, ...\n",
            "Connecting to objects.githubusercontent.com (objects.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 143099649 (136M) [application/octet-stream]\n",
            "Saving to: ‘yolov7x.pt’\n",
            "\n",
            "yolov7x.pt          100%[===================>] 136.47M  51.2MB/s    in 2.7s    \n",
            "\n",
            "2023-05-28 14:45:14 (51.2 MB/s) - ‘yolov7x.pt’ saved [143099649/143099649]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Before training, you neet to go to `yolov7/data/coco.yaml` and define your number of class, class names and train-val paths like that:\n",
        "\n",
        "```\n",
        "# COCO 2017 dataset http://cocodataset.org\n",
        "\n",
        "train: ../drone_dataset/train\n",
        "val: ../drone_dataset/valid\n",
        "\n",
        "# number of classes\n",
        "nc: 1\n",
        "\n",
        "# class names\n",
        "names: ['drone']\n",
        "```"
      ],
      "metadata": {
        "id": "Qzy2B3ueTCf7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "*   `!python train.py`: This is the command to execute the Python script \"train.py\" for training the YOLOv7 model.\n",
        "\n",
        "*  `--device 0`: This parameter specifies the device (GPU) to be used for training. In this case, it is set to device 0, indicating the first GPU device.\n",
        "\n",
        "*  `--batch-size 16`: This parameter determines the number of images in each batch during training. A batch size of 16 means that the model will process 16 images at a time before updating the weights.\n",
        "\n",
        "*  `--data data/coco.yaml`: This parameter specifies the path to the YAML file containing the dataset configuration. In this case, **the \"coco.yaml\" file is used, which provides information about the dataset, including the classes and paths to the training and validation data.**\n",
        "\n",
        "*  `--img 640 640`: This parameter sets the input image size for the model. The YOLOv7 model requires square input images, and here the dimensions are set to 640x640 pixels.\n",
        "\n",
        "*  `--epochs 32`: This parameter defines the number of epochs, which represents the number of times the entire training dataset will be passed through the model during training. In this case, the model will be trained for 32 epochs.\n",
        "\n",
        "*  `--weights yolov7x.pt`: This parameter specifies the initial weights of the model. The \"yolov7x.pt\" file contains the pre-trained weights for the YOLOv7 model, which will be used as the starting point for training.\n",
        "\n",
        "*  `--hyp data/hyp.scratch.p5.yaml`: This parameter indicates the path to the YAML file containing hyperparameters for training. Hyperparameters include learning rate, weight decay, and other settings that affect the training process. Here, the \"hyp.scratch.p5.yaml\" file is used.\n",
        "\n",
        "*  `--name yolov7x`: This parameter sets the name of the model during training. The name can be customized, and in this case, it is set to \"yolov7x\".\n",
        "\n"
      ],
      "metadata": {
        "id": "RPr-LxU9UJCG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "If you are using GPU, try this:"
      ],
      "metadata": {
        "id": "Z_9PdmaMP5cw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!python train.py --device 0 --batch-size 16 --data data/coco.yaml --img 640 640 --epochs 32 --weights yolov7x.pt --hyp data/hyp.scratch.p5.yaml --name yolov7x"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yrN77YHAcMpV",
        "outputId": "47449251-4a4d-40e3-9e7a-8d52d29bb147"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2023-05-28 14:57:07.828768: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
            "To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
            "2023-05-28 14:57:09.003739: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
            "YOLOR 🚀 v0.1-122-g3b41c2c torch 2.0.1+cu118 CUDA:0 (Tesla T4, 15101.8125MB)\n",
            "\n",
            "Namespace(weights='yolov7x.pt', cfg='', data='data/coco.yaml', hyp='data/hyp.scratch.p5.yaml', epochs=32, batch_size=16, img_size=[640, 640], rect=False, resume=False, nosave=False, notest=False, noautoanchor=False, evolve=False, bucket='', cache_images=False, image_weights=False, device='0', multi_scale=False, single_cls=False, adam=False, sync_bn=False, local_rank=-1, workers=8, project='runs/train', entity=None, name='yolov7x', exist_ok=False, quad=False, linear_lr=False, label_smoothing=0.0, upload_dataset=False, bbox_interval=-1, save_period=-1, artifact_alias='latest', freeze=[0], v5_metric=False, world_size=1, global_rank=-1, save_dir='runs/train/yolov7x2', total_batch_size=16)\n",
            "\u001b[34m\u001b[1mtensorboard: \u001b[0mStart with 'tensorboard --logdir runs/train', view at http://localhost:6006/\n",
            "\u001b[34m\u001b[1mhyperparameters: \u001b[0mlr0=0.01, lrf=0.1, momentum=0.937, weight_decay=0.0005, warmup_epochs=3.0, warmup_momentum=0.8, warmup_bias_lr=0.1, box=0.05, cls=0.3, cls_pw=1.0, obj=0.7, obj_pw=1.0, iou_t=0.2, anchor_t=4.0, fl_gamma=0.0, hsv_h=0.015, hsv_s=0.7, hsv_v=0.4, degrees=0.0, translate=0.2, scale=0.9, shear=0.0, perspective=0.0, flipud=0.0, fliplr=0.5, mosaic=1.0, mixup=0.15, copy_paste=0.0, paste_in=0.15, loss_ota=1\n",
            "\u001b[34m\u001b[1mwandb: \u001b[0mInstall Weights & Biases for YOLOR logging with 'pip install wandb' (recommended)\n",
            "Overriding model.yaml nc=80 with nc=1\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "  0                -1  1      1160  models.common.Conv                      [3, 40, 3, 1]                 \n",
            "  1                -1  1     28960  models.common.Conv                      [40, 80, 3, 2]                \n",
            "  2                -1  1     57760  models.common.Conv                      [80, 80, 3, 1]                \n",
            "  3                -1  1    115520  models.common.Conv                      [80, 160, 3, 2]               \n",
            "  4                -1  1     10368  models.common.Conv                      [160, 64, 1, 1]               \n",
            "  5                -2  1     10368  models.common.Conv                      [160, 64, 1, 1]               \n",
            "  6                -1  1     36992  models.common.Conv                      [64, 64, 3, 1]                \n",
            "  7                -1  1     36992  models.common.Conv                      [64, 64, 3, 1]                \n",
            "  8                -1  1     36992  models.common.Conv                      [64, 64, 3, 1]                \n",
            "  9                -1  1     36992  models.common.Conv                      [64, 64, 3, 1]                \n",
            " 10                -1  1     36992  models.common.Conv                      [64, 64, 3, 1]                \n",
            " 11                -1  1     36992  models.common.Conv                      [64, 64, 3, 1]                \n",
            " 12[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  1    103040  models.common.Conv                      [320, 320, 1, 1]              \n",
            " 14                -1  1         0  models.common.MP                        []                            \n",
            " 15                -1  1     51520  models.common.Conv                      [320, 160, 1, 1]              \n",
            " 16                -3  1     51520  models.common.Conv                      [320, 160, 1, 1]              \n",
            " 17                -1  1    230720  models.common.Conv                      [160, 160, 3, 2]              \n",
            " 18          [-1, -3]  1         0  models.common.Concat                    [1]                           \n",
            " 19                -1  1     41216  models.common.Conv                      [320, 128, 1, 1]              \n",
            " 20                -2  1     41216  models.common.Conv                      [320, 128, 1, 1]              \n",
            " 21                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 22                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 23                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 24                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 25                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 26                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 27[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            " 28                -1  1    410880  models.common.Conv                      [640, 640, 1, 1]              \n",
            " 29                -1  1         0  models.common.MP                        []                            \n",
            " 30                -1  1    205440  models.common.Conv                      [640, 320, 1, 1]              \n",
            " 31                -3  1    205440  models.common.Conv                      [640, 320, 1, 1]              \n",
            " 32                -1  1    922240  models.common.Conv                      [320, 320, 3, 2]              \n",
            " 33          [-1, -3]  1         0  models.common.Concat                    [1]                           \n",
            " 34                -1  1    164352  models.common.Conv                      [640, 256, 1, 1]              \n",
            " 35                -2  1    164352  models.common.Conv                      [640, 256, 1, 1]              \n",
            " 36                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 37                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 38                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 39                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 40                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 41                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 42[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            " 43                -1  1   1640960  models.common.Conv                      [1280, 1280, 1, 1]            \n",
            " 44                -1  1         0  models.common.MP                        []                            \n",
            " 45                -1  1    820480  models.common.Conv                      [1280, 640, 1, 1]             \n",
            " 46                -3  1    820480  models.common.Conv                      [1280, 640, 1, 1]             \n",
            " 47                -1  1   3687680  models.common.Conv                      [640, 640, 3, 2]              \n",
            " 48          [-1, -3]  1         0  models.common.Concat                    [1]                           \n",
            " 49                -1  1    328192  models.common.Conv                      [1280, 256, 1, 1]             \n",
            " 50                -2  1    328192  models.common.Conv                      [1280, 256, 1, 1]             \n",
            " 51                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 52                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 53                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 54                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 55                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 56                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 57[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            " 58                -1  1   1640960  models.common.Conv                      [1280, 1280, 1, 1]            \n",
            " 59                -1  1  11887360  models.common.SPPCSPC                   [1280, 640, 1]                \n",
            " 60                -1  1    205440  models.common.Conv                      [640, 320, 1, 1]              \n",
            " 61                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 62                43  1    410240  models.common.Conv                      [1280, 320, 1, 1]             \n",
            " 63          [-1, -2]  1         0  models.common.Concat                    [1]                           \n",
            " 64                -1  1    164352  models.common.Conv                      [640, 256, 1, 1]              \n",
            " 65                -2  1    164352  models.common.Conv                      [640, 256, 1, 1]              \n",
            " 66                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 67                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 68                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 69                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 70                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 71                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 72[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            " 73                -1  1    410240  models.common.Conv                      [1280, 320, 1, 1]             \n",
            " 74                -1  1     51520  models.common.Conv                      [320, 160, 1, 1]              \n",
            " 75                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 76                28  1    102720  models.common.Conv                      [640, 160, 1, 1]              \n",
            " 77          [-1, -2]  1         0  models.common.Concat                    [1]                           \n",
            " 78                -1  1     41216  models.common.Conv                      [320, 128, 1, 1]              \n",
            " 79                -2  1     41216  models.common.Conv                      [320, 128, 1, 1]              \n",
            " 80                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 81                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 82                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 83                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 84                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 85                -1  1    147712  models.common.Conv                      [128, 128, 3, 1]              \n",
            " 86[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            " 87                -1  1    102720  models.common.Conv                      [640, 160, 1, 1]              \n",
            " 88                -1  1         0  models.common.MP                        []                            \n",
            " 89                -1  1     25920  models.common.Conv                      [160, 160, 1, 1]              \n",
            " 90                -3  1     25920  models.common.Conv                      [160, 160, 1, 1]              \n",
            " 91                -1  1    230720  models.common.Conv                      [160, 160, 3, 2]              \n",
            " 92      [-1, -3, 73]  1         0  models.common.Concat                    [1]                           \n",
            " 93                -1  1    164352  models.common.Conv                      [640, 256, 1, 1]              \n",
            " 94                -2  1    164352  models.common.Conv                      [640, 256, 1, 1]              \n",
            " 95                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 96                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 97                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 98                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            " 99                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            "100                -1  1    590336  models.common.Conv                      [256, 256, 3, 1]              \n",
            "101[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            "102                -1  1    410240  models.common.Conv                      [1280, 320, 1, 1]             \n",
            "103                -1  1         0  models.common.MP                        []                            \n",
            "104                -1  1    103040  models.common.Conv                      [320, 320, 1, 1]              \n",
            "105                -3  1    103040  models.common.Conv                      [320, 320, 1, 1]              \n",
            "106                -1  1    922240  models.common.Conv                      [320, 320, 3, 2]              \n",
            "107      [-1, -3, 59]  1         0  models.common.Concat                    [1]                           \n",
            "108                -1  1    656384  models.common.Conv                      [1280, 512, 1, 1]             \n",
            "109                -2  1    656384  models.common.Conv                      [1280, 512, 1, 1]             \n",
            "110                -1  1   2360320  models.common.Conv                      [512, 512, 3, 1]              \n",
            "111                -1  1   2360320  models.common.Conv                      [512, 512, 3, 1]              \n",
            "112                -1  1   2360320  models.common.Conv                      [512, 512, 3, 1]              \n",
            "113                -1  1   2360320  models.common.Conv                      [512, 512, 3, 1]              \n",
            "114                -1  1   2360320  models.common.Conv                      [512, 512, 3, 1]              \n",
            "115                -1  1   2360320  models.common.Conv                      [512, 512, 3, 1]              \n",
            "116[-1, -3, -5, -7, -8]  1         0  models.common.Concat                    [1]                           \n",
            "117                -1  1   1639680  models.common.Conv                      [2560, 640, 1, 1]             \n",
            "118                87  1    461440  models.common.Conv                      [160, 320, 3, 1]              \n",
            "119               102  1   1844480  models.common.Conv                      [320, 640, 3, 1]              \n",
            "120               117  1   7375360  models.common.Conv                      [640, 1280, 3, 1]             \n",
            "121   [118, 119, 120]  1     40374  models.yolo.Detect                      [1, [[12, 16, 19, 36, 40, 28], [36, 75, 76, 55, 72, 146], [142, 110, 192, 243, 459, 401]], [320, 640, 1280]]\n",
            "Model Summary: 459 layers, 70812798 parameters, 70812798 gradients\n",
            "\n",
            "Transferred 632/638 items from yolov7x.pt\n",
            "Scaled weight_decay = 0.0005\n",
            "Optimizer groups: 108 .bias, 108 conv.weight, 105 other\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mScanning '../drone_dataset/train/labels.cache' images and labels... 1012 found, 0 missing, 0 empty, 0 corrupted: 100% 1012/1012 [00:00<?, ?it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mScanning '../drone_dataset/valid/labels.cache' images and labels... 347 found, 0 missing, 0 empty, 0 corrupted: 100% 347/347 [00:00<?, ?it/s]\n",
            "\n",
            "\u001b[34m\u001b[1mautoanchor: \u001b[0mAnalyzing anchors... anchors/target = 3.11, Best Possible Recall (BPR) = 1.0000\n",
            "Image sizes 640 train, 640 test\n",
            "Using 2 dataloader workers\n",
            "Logging results to runs/train/yolov7x2\n",
            "Starting training for 32 epochs...\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      0/31     3.45G   0.07819   0.02026         0   0.09845        60       640:   2% 1/64 [00:22<23:25, 22.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31      7.4G   0.07823   0.02011         0   0.09835        56       640:   8% 5/64 [00:29<03:31,  3.59s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31       12G   0.07906    0.0196         0   0.09866        56       640:  14% 9/64 [00:37<02:07,  2.32s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07847   0.01931         0   0.09778        51       640:  17% 11/64 [00:42<02:13,  2.53s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07743   0.01923         0   0.09666        45       640:  19% 12/64 [00:44<01:51,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07752   0.01928         0    0.0968        70       640:  20% 13/64 [00:46<01:54,  2.25s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07739   0.01918         0   0.09657        47       640:  22% 14/64 [00:47<01:39,  1.99s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07657   0.01895         0   0.09552        55       640:  28% 18/64 [00:55<01:26,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07571    0.0188         0    0.0945        45       640:  33% 21/64 [01:02<01:34,  2.20s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07559   0.01871         0    0.0943        47       640:  34% 22/64 [01:04<01:25,  2.03s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07503   0.01854         0   0.09357        34       640:  38% 24/64 [01:08<01:27,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07467   0.01839         0   0.09306        46       640:  41% 26/64 [01:13<01:21,  2.13s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07425    0.0183         0   0.09256        37       640:  42% 27/64 [01:15<01:23,  2.25s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07373   0.01806         0   0.09179        46       640:  48% 31/64 [01:24<01:12,  2.19s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07361   0.01801         0   0.09162        54       640:  50% 32/64 [01:25<01:00,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07193   0.01765         0   0.08959        52       640:  66% 42/64 [01:47<00:48,  2.21s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.07123   0.01749         0   0.08872        49       640:  72% 46/64 [01:55<00:37,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G    0.0704   0.01743         0   0.08784        64       640:  78% 50/64 [02:04<00:30,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.06945    0.0172         0   0.08665        49       640:  91% 58/64 [02:22<00:13,  2.25s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.06919   0.01712         0   0.08631        58       640:  94% 60/64 [02:26<00:07,  1.97s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.06912   0.01707         0   0.08618        42       640:  95% 61/64 [02:30<00:07,  2.58s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     14.2G   0.06879   0.01702         0   0.08581        48       640:  98% 63/64 [02:34<00:02,  2.34s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      0/31     2.38G   0.06842   0.01705         0   0.08547        14       640: 100% 64/64 [02:43<00:00,  2.55s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:   0% 0/11 [00:00<?, ?it/s]/usr/local/lib/python3.10/dist-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3483.)\n",
            "  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:40<00:00,  3.72s/it]\n",
            "                 all         347         369       0.137       0.228      0.0886      0.0177\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      1/31     9.94G   0.05963   0.01576         0   0.07539        57       640:  12% 8/64 [00:12<01:40,  1.79s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05929    0.0154         0   0.07469        54       640:  19% 12/64 [00:19<01:28,  1.70s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05879   0.01539         0   0.07418        58       640:  23% 15/64 [00:27<01:55,  2.36s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     8.06G   0.05793   0.01529         0   0.07322        40       640:  25% 16/64 [00:28<01:40,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05661   0.01504         0   0.07166        37       640:  28% 18/64 [00:32<01:34,  2.05s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05697   0.01485         0   0.07182        48       640:  36% 23/64 [00:43<01:19,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05664   0.01474         0   0.07137        55       640:  39% 25/64 [00:47<01:14,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05715   0.01473         0   0.07187        62       640:  52% 33/64 [01:04<01:02,  2.03s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05695   0.01456         0   0.07151        49       640:  58% 37/64 [01:12<00:50,  1.88s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05674   0.01445         0   0.07119        44       640:  61% 39/64 [01:15<00:42,  1.69s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05646   0.01431         0   0.07077        61       640:  69% 44/64 [01:26<00:47,  2.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05689   0.01425         0   0.07114        56       640:  77% 49/64 [01:36<00:31,  2.12s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05706   0.01413         0   0.07119        57       640:  83% 53/64 [01:44<00:20,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05714   0.01411         0   0.07125        59       640:  84% 54/64 [01:48<00:26,  2.68s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05738   0.01404         0   0.07142        71       640:  92% 59/64 [01:59<00:11,  2.33s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      1/31     14.3G   0.05727   0.01381         0   0.07108         6       640: 100% 64/64 [02:08<00:00,  2.00s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.34s/it]\n",
            "                 all         347         369       0.505       0.493       0.463       0.168\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      2/31     14.2G   0.05664   0.01305         0    0.0697        47       640:  11% 7/64 [00:09<01:16,  1.34s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06023   0.01273         0   0.07295        49       640:  17% 11/64 [00:18<01:35,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G    0.0605   0.01245         0   0.07295        61       640:  23% 15/64 [00:27<01:45,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06095   0.01228         0   0.07324        48       640:  30% 19/64 [00:35<01:31,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06134   0.01209         0   0.07344        44       640:  34% 22/64 [00:42<01:25,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06144   0.01205         0   0.07349        55       640:  36% 23/64 [00:43<01:14,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06092   0.01187         0   0.07279        63       640:  44% 28/64 [00:56<01:18,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06126   0.01186         0   0.07312        74       640:  45% 29/64 [00:57<01:07,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06184   0.01177         0   0.07361        57       640:  58% 37/64 [01:13<00:51,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06147   0.01173         0    0.0732        48       640:  61% 39/64 [01:18<00:49,  1.97s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06097   0.01164         0   0.07261        48       640:  69% 44/64 [01:28<00:41,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06078   0.01156         0   0.07234        68       640:  83% 53/64 [01:46<00:19,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G    0.0608   0.01155         0   0.07235        58       640:  86% 55/64 [01:50<00:17,  1.95s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      2/31     14.2G   0.06127   0.01151         0   0.07278        16       640: 100% 64/64 [02:08<00:00,  2.01s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.35s/it]\n",
            "                 all         347         369       0.424       0.499        0.43       0.155\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      3/31     14.2G   0.06183    0.0114         0   0.07323        51       640:  20% 13/64 [00:21<01:34,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.06125   0.01136         0   0.07261        47       640:  23% 15/64 [00:25<01:28,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05695   0.01119         0   0.06813        52       640:  33% 21/64 [00:37<01:18,  1.84s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05676   0.01131         0   0.06807        70       640:  34% 22/64 [00:39<01:27,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05725   0.01124         0   0.06849        55       640:  44% 28/64 [00:51<01:06,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05713   0.01115         0   0.06828        51       640:  48% 31/64 [00:56<00:56,  1.73s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05747   0.01113         0    0.0686        61       640:  56% 36/64 [01:06<00:51,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05823   0.01111         0   0.06934        45       640:  72% 46/64 [01:25<00:35,  1.99s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G    0.0586   0.01108         0   0.06968        48       640:  78% 50/64 [01:33<00:27,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05849   0.01106         0   0.06955        57       640:  81% 52/64 [01:38<00:27,  2.26s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05802   0.01099         0   0.06901        42       640:  84% 54/64 [01:42<00:20,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05794    0.0109         0   0.06884        48       640:  92% 59/64 [01:56<00:12,  2.43s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      3/31     14.2G   0.05841   0.01078         0   0.06919         9       640: 100% 64/64 [02:04<00:00,  1.94s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:06<00:03,  1.01it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.20s/it]\n",
            "                 all         347         369       0.654        0.46       0.513       0.223\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      4/31     14.2G   0.05126   0.01024         0    0.0615        60       640:  23% 15/64 [00:26<01:31,  1.87s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05181   0.01021         0   0.06202        45       640:  27% 17/64 [00:29<01:24,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05132   0.01006         0   0.06139        51       640:  30% 19/64 [00:32<01:14,  1.65s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05398   0.01016         0   0.06414        56       640:  34% 22/64 [00:39<01:30,  2.14s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05494   0.01011         0   0.06505        42       640:  36% 23/64 [00:41<01:19,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05803   0.01023         0   0.06826        52       640:  48% 31/64 [00:56<00:59,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05654   0.01023         0   0.06677        54       640:  52% 33/64 [01:01<01:02,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05575   0.01024         0   0.06599        53       640:  53% 34/64 [01:03<01:02,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05598   0.01017         0   0.06615        37       640:  62% 40/64 [01:14<00:45,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05609   0.01019         0   0.06629        59       640:  66% 42/64 [01:19<00:49,  2.23s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05623   0.01018         0    0.0664        65       640:  70% 45/64 [01:24<00:36,  1.94s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05558   0.01011         0    0.0657        45       640:  83% 53/64 [01:40<00:19,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05581    0.0101         0   0.06592        43       640:  89% 57/64 [01:49<00:14,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05613   0.01012         0   0.06625        67       640:  92% 59/64 [01:55<00:12,  2.47s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05619   0.01011         0    0.0663        40       640:  94% 60/64 [01:58<00:09,  2.45s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05609   0.01006         0   0.06616        32       640:  95% 61/64 [01:59<00:06,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      4/31     14.2G   0.05605   0.01011         0   0.06616        26       640: 100% 64/64 [02:03<00:00,  1.94s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.25s/it]\n",
            "                 all         347         369       0.422       0.393       0.368       0.147\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      5/31     14.2G   0.05858   0.01154         0   0.07011        50       640:   9% 6/64 [00:07<01:19,  1.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05765   0.01061         0   0.06826        49       640:  16% 10/64 [00:15<01:26,  1.60s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05768    0.0106         0   0.06827        62       640:  17% 11/64 [00:16<01:25,  1.61s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05691   0.01062         0   0.06752        52       640:  20% 13/64 [00:20<01:26,  1.69s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05488   0.01055         0   0.06542        49       640:  23% 15/64 [00:24<01:27,  1.79s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G    0.0554    0.0105         0    0.0659        39       640:  27% 17/64 [00:28<01:30,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05689   0.01077         0   0.06766        61       640:  33% 21/64 [00:37<01:33,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05611   0.01071         0   0.06681        55       640:  36% 23/64 [00:41<01:18,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05646   0.01081         0   0.06727        56       640:  39% 25/64 [00:46<01:22,  2.10s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05534   0.01069         0   0.06603        37       640:  41% 26/64 [00:49<01:35,  2.51s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05366    0.0107         0   0.06435        50       640:  47% 30/64 [00:55<01:01,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05362   0.01067         0   0.06429        45       640:  48% 31/64 [00:57<00:54,  1.65s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05496   0.01068         0   0.06565        62       640:  70% 45/64 [01:24<00:33,  1.75s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05527   0.01058         0   0.06585        43       640:  83% 53/64 [01:39<00:20,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05541   0.01059         0   0.06601        47       640:  86% 55/64 [01:44<00:18,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05509   0.01046         0   0.06555        42       640:  95% 61/64 [01:58<00:06,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      5/31     14.2G   0.05466   0.01037         0   0.06503         8       640: 100% 64/64 [02:02<00:00,  1.92s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  36% 4/11 [00:03<00:06,  1.02it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:12<00:00,  1.10s/it]\n",
            "                 all         347         369       0.749       0.518       0.588       0.225\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      6/31     14.2G   0.04966  0.009493         0   0.05915        50       640:   6% 4/64 [00:04<01:10,  1.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05522   0.01021         0   0.06543        58       640:  19% 12/64 [00:19<01:30,  1.74s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05239   0.01041         0    0.0628        53       640:  23% 15/64 [00:26<01:30,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.04935   0.01036         0   0.05971        46       640:  27% 17/64 [00:29<01:26,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05144   0.01023         0   0.06167        62       640:  36% 23/64 [00:42<01:16,  1.88s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05102   0.01014         0   0.06116        36       640:  41% 26/64 [00:50<01:28,  2.34s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05138   0.01003         0   0.06141        53       640:  45% 29/64 [00:54<01:01,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G      0.05  0.009966         0   0.05996        59       640:  56% 36/64 [01:06<00:45,  1.61s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.04989   0.01024         0   0.06013        52       640:  69% 44/64 [01:23<00:47,  2.39s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05019   0.01023         0   0.06042        47       640:  70% 45/64 [01:25<00:39,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05092   0.01028         0    0.0612        59       640:  73% 47/64 [01:30<00:36,  2.14s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05143   0.01026         0   0.06169        47       640:  77% 49/64 [01:33<00:27,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G    0.0519   0.01029         0   0.06219        76       640:  80% 51/64 [01:36<00:23,  1.79s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05227    0.0103         0   0.06257        42       640:  91% 58/64 [01:55<00:16,  2.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05213   0.01028         0   0.06241        51       640:  95% 61/64 [01:59<00:05,  1.91s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      6/31     14.2G   0.05204    0.0103         0   0.06234        17       640: 100% 64/64 [02:04<00:00,  1.94s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.21s/it]\n",
            "                 all         347         369       0.727       0.439       0.524       0.195\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      7/31     14.2G   0.05609   0.01016         0   0.06625        68       640:  12% 8/64 [00:11<01:23,  1.48s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05574    0.0101         0   0.06584        43       640:  14% 9/64 [00:14<01:45,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G    0.0579   0.01021         0   0.06812        59       640:  16% 10/64 [00:15<01:35,  1.77s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05652   0.01001         0   0.06653        40       640:  17% 11/64 [00:17<01:37,  1.84s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05472  0.009867         0   0.06459        38       640:  19% 12/64 [00:19<01:32,  1.77s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G    0.0537    0.0101         0    0.0638        45       640:  22% 14/64 [00:22<01:25,  1.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05415   0.01001         0   0.06416        40       640:  25% 16/64 [00:27<01:31,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05342   0.00999         0   0.06341        48       640:  30% 19/64 [00:35<01:40,  2.24s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05243  0.009837         0   0.06227        52       640:  33% 21/64 [00:38<01:21,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05328   0.01005         0   0.06333        78       640:  38% 24/64 [00:44<01:25,  2.14s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05351   0.01002         0   0.06353        53       640:  39% 25/64 [00:47<01:28,  2.27s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05238    0.0099         0   0.06228        48       640:  58% 37/64 [01:11<01:03,  2.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05282  0.009887         0    0.0627        54       640:  59% 38/64 [01:12<00:52,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05351  0.009843         0   0.06335        49       640:  67% 43/64 [01:21<00:40,  1.91s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05311  0.009827         0   0.06293        57       640:  72% 46/64 [01:27<00:34,  1.91s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05286  0.009671         0   0.06253        63       640:  84% 54/64 [01:41<00:16,  1.65s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      7/31     14.2G   0.05325  0.009536         0   0.06279         6       640: 100% 64/64 [02:03<00:00,  1.93s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:06<00:03,  1.07it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.28s/it]\n",
            "                 all         347         369       0.711       0.515       0.582        0.24\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      8/31     14.2G   0.04651  0.009303         0   0.05581        49       640:   8% 5/64 [00:06<01:21,  1.38s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.04358  0.009165         0   0.05274        43       640:   9% 6/64 [00:07<01:16,  1.33s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05233  0.009662         0   0.06199        43       640:  22% 14/64 [00:23<01:35,  1.91s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05189  0.009407         0    0.0613        57       640:  33% 21/64 [00:35<01:16,  1.77s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05077  0.009563         0   0.06033        46       640:  38% 24/64 [00:41<01:17,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05067  0.009664         0   0.06033        46       640:  50% 32/64 [00:57<00:58,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05121  0.009697         0    0.0609        67       640:  53% 34/64 [01:02<01:00,  2.03s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05121   0.00971         0   0.06092        35       640:  59% 38/64 [01:10<00:49,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05147  0.009721         0   0.06119        52       640:  62% 40/64 [01:15<00:49,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05096  0.009738         0    0.0607        54       640:  64% 41/64 [01:16<00:46,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05106  0.009722         0   0.06078        50       640:  66% 42/64 [01:18<00:38,  1.75s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05132  0.009708         0   0.06103        48       640:  67% 43/64 [01:20<00:43,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05168  0.009746         0   0.06143        46       640:  72% 46/64 [01:26<00:35,  1.97s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05292  0.009719         0   0.06263        53       640:  91% 58/64 [01:51<00:12,  2.05s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05322  0.009725         0   0.06295        78       640:  94% 60/64 [01:56<00:08,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      8/31     14.2G   0.05332  0.009782         0    0.0631        17       640: 100% 64/64 [02:02<00:00,  1.92s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  45% 5/11 [00:04<00:05,  1.08it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:07<00:04,  1.13s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.24s/it]\n",
            "                 all         347         369       0.656       0.558        0.58       0.247\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "      9/31     14.2G    0.0309  0.008889         0   0.03978        40       640:   2% 1/64 [00:01<01:14,  1.19s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.04875  0.008632         0   0.05739        36       640:   8% 5/64 [00:05<01:08,  1.16s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.04808  0.008711         0   0.05679        42       640:  11% 7/64 [00:08<01:12,  1.27s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.04922  0.009549         0   0.05877        56       640:  17% 11/64 [00:17<01:33,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05134   0.00959         0   0.06093        56       640:  23% 15/64 [00:23<01:19,  1.63s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G      0.05  0.009721         0   0.05972        43       640:  30% 19/64 [00:31<01:17,  1.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.04972  0.009664         0   0.05938        56       640:  34% 22/64 [00:40<01:49,  2.62s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.04984    0.0097         0   0.05954        53       640:  36% 23/64 [00:41<01:29,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05327  0.009834         0    0.0631        65       640:  66% 42/64 [01:15<00:40,  1.84s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05317  0.009823         0   0.06299        36       640:  69% 44/64 [01:20<00:45,  2.25s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05324  0.009839         0   0.06308        50       640:  70% 45/64 [01:21<00:37,  1.95s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05406  0.009797         0   0.06386        48       640:  77% 49/64 [01:28<00:24,  1.62s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05414  0.009822         0   0.06397        53       640:  83% 53/64 [01:37<00:22,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05437  0.009863         0   0.06423        62       640:  86% 55/64 [01:42<00:18,  2.10s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05445  0.009885         0   0.06433        61       640:  89% 57/64 [01:48<00:17,  2.52s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05448  0.009885         0   0.06437        55       640:  91% 58/64 [01:52<00:16,  2.77s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05443  0.009897         0   0.06433        55       640:  92% 59/64 [01:53<00:11,  2.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "      9/31     14.2G   0.05449  0.009963         0   0.06445        22       640: 100% 64/64 [02:03<00:00,  1.92s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.26s/it]\n",
            "                 all         347         369       0.424       0.347       0.332       0.117\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "  0% 0/64 [00:00<?, ?it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G     0.051   0.01013         0   0.06113        56       640:  12% 8/64 [00:10<01:08,  1.22s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G    0.0512   0.01008         0   0.06128        45       640:  19% 12/64 [00:18<01:31,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05412   0.01022         0   0.06434        59       640:  23% 15/64 [00:24<01:31,  1.88s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05483   0.01017         0   0.06501        48       640:  28% 18/64 [00:31<01:38,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05337   0.01009         0   0.06346        52       640:  36% 23/64 [00:39<01:09,  1.71s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05183   0.01016         0   0.06198        56       640:  58% 37/64 [01:08<00:46,  1.71s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05162   0.01021         0   0.06183        51       640:  62% 40/64 [01:14<00:40,  1.71s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G    0.0518   0.01023         0   0.06203        62       640:  64% 41/64 [01:15<00:36,  1.58s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05155   0.01024         0   0.06179        38       640:  70% 45/64 [01:22<00:32,  1.69s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05177   0.01022         0   0.06199        54       640:  72% 46/64 [01:24<00:34,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05232   0.01015         0   0.06247        46       640:  80% 51/64 [01:35<00:27,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05234   0.01014         0   0.06248        58       640:  81% 52/64 [01:37<00:25,  2.16s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05223   0.01013         0   0.06236        53       640:  84% 54/64 [01:42<00:23,  2.39s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G   0.05245   0.01019         0   0.06263        50       640:  92% 59/64 [01:55<00:11,  2.35s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     10/31     14.2G    0.0525   0.01012         0   0.06262        11       640: 100% 64/64 [02:07<00:00,  1.99s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:05<00:03,  1.10it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:12<00:00,  1.12s/it]\n",
            "                 all         347         369       0.444       0.377       0.352       0.164\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     11/31     14.2G   0.03793  0.009262         0   0.04719        53       640:   3% 2/64 [00:02<01:21,  1.32s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.04501  0.009397         0   0.05441        49       640:  11% 7/64 [00:10<01:42,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G    0.0464   0.01025         0   0.05665        64       640:  16% 10/64 [00:17<01:55,  2.14s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.04995  0.009801         0   0.05975        60       640:  22% 14/64 [00:25<01:28,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.05293   0.01013         0   0.06305        57       640:  25% 16/64 [00:29<01:29,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.05143  0.009983         0   0.06141        43       640:  33% 21/64 [00:39<01:21,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.05294   0.01001         0   0.06295        57       640:  38% 24/64 [00:45<01:18,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G    0.0514  0.009977         0   0.06137        51       640:  47% 30/64 [00:54<00:54,  1.60s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G    0.0511  0.009795         0   0.06089        51       640:  56% 36/64 [01:05<00:44,  1.60s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.04939  0.009477         0   0.05887        55       640:  72% 46/64 [01:25<00:37,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G   0.05067  0.009758         0   0.06043        48       640:  97% 62/64 [02:00<00:04,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     11/31     14.2G    0.0505  0.009779         0   0.06028        14       640: 100% 64/64 [02:05<00:00,  1.96s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.28s/it]\n",
            "                 all         347         369       0.555       0.271       0.301       0.129\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     12/31     14.3G   0.05552   0.01111         0   0.06664        50       640:   3% 2/64 [00:02<01:24,  1.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05393   0.01137         0    0.0653        63       640:   5% 3/64 [00:04<01:25,  1.40s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05915   0.01073         0   0.06989        46       640:   6% 4/64 [00:05<01:21,  1.36s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05852   0.01037         0   0.06889        40       640:   8% 5/64 [00:06<01:17,  1.32s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05276   0.01013         0   0.06289        38       640:  11% 7/64 [00:12<02:02,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05422  0.009812         0   0.06403        44       640:  19% 12/64 [00:22<01:44,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05323  0.009804         0   0.06303        62       640:  22% 14/64 [00:25<01:33,  1.87s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05456  0.009734         0   0.06429        51       640:  25% 16/64 [00:29<01:26,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05494  0.009657         0    0.0646        44       640:  28% 18/64 [00:34<01:32,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05378  0.009519         0    0.0633        43       640:  36% 23/64 [00:41<01:02,  1.52s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G    0.0537  0.009482         0   0.06318        53       640:  38% 24/64 [00:46<01:47,  2.70s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05202  0.009576         0    0.0616        64       640:  48% 31/64 [01:00<01:07,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05236  0.009633         0     0.062        40       640:  61% 39/64 [01:14<00:41,  1.66s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05293  0.009723         0   0.06265        58       640:  69% 44/64 [01:26<00:44,  2.22s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05236  0.009787         0   0.06215        45       640:  73% 47/64 [01:31<00:32,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05182    0.0098         0   0.06162        43       640:  75% 48/64 [01:34<00:36,  2.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05326  0.009877         0   0.06314        55       640:  88% 56/64 [01:52<00:20,  2.55s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05353  0.009948         0   0.06348        52       640:  92% 59/64 [01:58<00:11,  2.21s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05361   0.01001         0   0.06362        56       640:  95% 61/64 [02:03<00:06,  2.25s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     12/31     14.3G   0.05341  0.009996         0   0.06341        17       640: 100% 64/64 [02:07<00:00,  2.00s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:06<00:04,  1.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.26s/it]\n",
            "                 all         347         369       0.417       0.434       0.342       0.134\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     13/31     14.3G   0.05179  0.009606         0   0.06139        59       640:  23% 15/64 [00:23<01:37,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05247  0.009625         0   0.06209        52       640:  25% 16/64 [00:24<01:27,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05354  0.009493         0   0.06303        39       640:  27% 17/64 [00:29<02:00,  2.56s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05424   0.00955         0   0.06379        62       640:  28% 18/64 [00:30<01:39,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05424  0.009619         0   0.06386        63       640:  30% 19/64 [00:32<01:42,  2.27s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05309  0.009602         0   0.06269        42       640:  31% 20/64 [00:34<01:27,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05341  0.009693         0    0.0631        46       640:  36% 23/64 [00:40<01:22,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05425  0.009769         0   0.06402        63       640:  38% 24/64 [00:42<01:22,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05439  0.009739         0   0.06413        55       640:  41% 26/64 [00:47<01:22,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05398  0.009702         0   0.06368        35       640:  47% 30/64 [00:55<01:07,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05366  0.009647         0    0.0633        46       640:  59% 38/64 [01:09<00:43,  1.68s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05363  0.009676         0   0.06331        48       640:  61% 39/64 [01:11<00:38,  1.53s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G    0.0534  0.009674         0   0.06308        48       640:  64% 41/64 [01:15<00:39,  1.73s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05384  0.009599         0   0.06343        44       640:  70% 45/64 [01:22<00:32,  1.73s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G    0.0543  0.009618         0   0.06392        63       640:  78% 50/64 [01:33<00:31,  2.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05402  0.009643         0   0.06366        37       640:  86% 55/64 [01:44<00:19,  2.19s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G    0.0541  0.009669         0   0.06377        53       640:  91% 58/64 [01:52<00:15,  2.58s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05425  0.009647         0    0.0639        48       640:  95% 61/64 [01:58<00:06,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     13/31     14.3G   0.05369  0.009542         0   0.06323         5       640: 100% 64/64 [02:02<00:00,  1.91s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.22s/it]\n",
            "                 all         347         369       0.443       0.453       0.336       0.145\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     14/31     14.3G     0.055   0.01116         0   0.06616        66       640:   6% 4/64 [00:04<01:12,  1.20s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05577   0.01031         0   0.06608        41       640:  11% 7/64 [00:08<01:09,  1.22s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05498   0.01035         0   0.06532        59       640:  12% 8/64 [00:12<01:51,  1.99s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05481   0.01048         0    0.0653        62       640:  14% 9/64 [00:13<01:36,  1.75s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G     0.055   0.01102         0   0.06602        57       640:  17% 11/64 [00:17<01:32,  1.75s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G    0.0548   0.01067         0   0.06547        47       640:  20% 13/64 [00:21<01:31,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05561   0.01087         0   0.06649        62       640:  23% 15/64 [00:24<01:29,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05452   0.01072         0   0.06524        48       640:  30% 19/64 [00:34<01:29,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G    0.0549   0.01043         0   0.06533        60       640:  34% 22/64 [00:41<01:27,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G    0.0523   0.01031         0   0.06261        51       640:  42% 27/64 [00:49<01:03,  1.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05157   0.01017         0   0.06174        37       640:  48% 31/64 [00:56<00:55,  1.67s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05158   0.01021         0   0.06179        52       640:  53% 34/64 [01:02<00:54,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05028   0.01028         0   0.06056        49       640:  56% 36/64 [01:07<01:02,  2.24s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05007   0.01005         0   0.06012        55       640:  69% 44/64 [01:24<00:49,  2.49s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05083   0.01001         0   0.06084        51       640:  77% 49/64 [01:33<00:30,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05103   0.01003         0   0.06107        55       640:  81% 52/64 [01:40<00:23,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05064   0.01006         0    0.0607        63       640:  86% 55/64 [01:45<00:17,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     14/31     14.3G   0.05012  0.009914         0   0.06003        10       640: 100% 64/64 [02:03<00:00,  1.93s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:12<00:00,  1.10s/it]\n",
            "                 all         347         369       0.645       0.415        0.47       0.215\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     15/31     14.3G   0.04853   0.00943         0   0.05796        56       640:   5% 3/64 [00:03<01:14,  1.23s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05377  0.009386         0   0.06316        57       640:  22% 14/64 [00:22<01:27,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05171  0.009397         0   0.06111        50       640:  23% 15/64 [00:25<01:50,  2.26s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05163  0.009381         0   0.06101        50       640:  25% 16/64 [00:27<01:37,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05184   0.00933         0   0.06117        44       640:  27% 17/64 [00:28<01:25,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05168  0.009729         0   0.06141        49       640:  34% 22/64 [00:39<01:33,  2.21s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05147   0.00968         0   0.06115        46       640:  36% 23/64 [00:40<01:19,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05119  0.009785         0   0.06098        55       640:  39% 25/64 [00:44<01:10,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05091  0.009687         0   0.06059        46       640:  45% 29/64 [00:53<01:15,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04991  0.009578         0   0.05949        67       640:  61% 39/64 [01:12<00:45,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04931  0.009592         0   0.05891        46       640:  69% 44/64 [01:23<00:36,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04928  0.009582         0   0.05887        53       640:  73% 47/64 [01:27<00:25,  1.51s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04941  0.009592         0     0.059        58       640:  75% 48/64 [01:31<00:36,  2.27s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04939  0.009733         0   0.05912        57       640:  83% 53/64 [01:40<00:20,  1.84s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04944  0.009714         0   0.05916        44       640:  86% 55/64 [01:45<00:19,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.04977  0.009656         0   0.05943        51       640:  89% 57/64 [01:50<00:14,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05003  0.009688         0   0.05972        44       640:  97% 62/64 [02:03<00:04,  2.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     15/31     14.3G   0.05019  0.009751         0   0.05995        26       640: 100% 64/64 [02:04<00:00,  1.95s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.23s/it]\n",
            "                 all         347         369       0.755       0.528       0.581       0.283\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     16/31     14.3G   0.05221  0.009221         0   0.06143        49       640:  11% 7/64 [00:09<01:17,  1.36s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.05002   0.00892         0   0.05894        42       640:  16% 10/64 [00:15<01:46,  1.97s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04966  0.008763         0   0.05842        39       640:  17% 11/64 [00:16<01:36,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.05144  0.008933         0   0.06037        62       640:  19% 12/64 [00:21<02:13,  2.58s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.05178  0.008965         0   0.06074        51       640:  20% 13/64 [00:22<01:50,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.05102  0.009009         0   0.06003        49       640:  23% 15/64 [00:26<01:33,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.05094  0.009081         0   0.06002        55       640:  27% 17/64 [00:30<01:34,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04775  0.008988         0   0.05674        38       640:  36% 23/64 [00:41<01:08,  1.67s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04754  0.009248         0   0.05679        70       640:  41% 26/64 [00:50<01:28,  2.32s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04852  0.009288         0   0.05781        60       640:  48% 31/64 [01:00<01:13,  2.21s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04854  0.009309         0   0.05785        60       640:  50% 32/64 [01:02<01:08,  2.14s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04748  0.009294         0   0.05678        40       640:  59% 38/64 [01:13<00:51,  1.97s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G    0.0477  0.009302         0     0.057        43       640:  70% 45/64 [01:25<00:32,  1.74s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04786  0.009357         0   0.05722        76       640:  78% 50/64 [01:34<00:26,  1.87s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G   0.04755  0.009452         0     0.057        77       640:  94% 60/64 [01:58<00:09,  2.41s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     16/31     14.3G    0.0477  0.009399         0    0.0571        12       640: 100% 64/64 [02:04<00:00,  1.94s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.22s/it]\n",
            "                 all         347         369        0.75        0.61       0.659       0.328\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     17/31     14.3G   0.04795  0.009533         0   0.05748        35       640:   9% 6/64 [00:07<01:14,  1.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04846  0.009716         0   0.05818        39       640:  16% 10/64 [00:15<01:33,  1.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04848  0.009917         0    0.0584        66       640:  17% 11/64 [00:19<02:07,  2.40s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04836  0.009829         0   0.05819        45       640:  20% 13/64 [00:22<01:46,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04871  0.009658         0   0.05837        45       640:  22% 14/64 [00:23<01:31,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.05024  0.009701         0   0.05994        50       640:  33% 21/64 [00:37<01:19,  1.84s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.05032  0.009729         0   0.06005        59       640:  34% 22/64 [00:39<01:26,  2.05s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G    0.0504  0.009683         0   0.06009        50       640:  36% 23/64 [00:41<01:18,  1.91s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04925  0.009609         0   0.05886        73       640:  58% 37/64 [01:09<00:47,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04923  0.009629         0   0.05886        62       640:  59% 38/64 [01:11<00:51,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04873  0.009611         0   0.05834        52       640:  70% 45/64 [01:25<00:38,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04862  0.009662         0   0.05828        64       640:  72% 46/64 [01:28<00:41,  2.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04948  0.009617         0    0.0591        57       640:  86% 55/64 [01:47<00:20,  2.24s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04907  0.009625         0    0.0587        62       640:  89% 57/64 [01:52<00:15,  2.23s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04942  0.009524         0   0.05895        38       640:  95% 61/64 [02:00<00:05,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     17/31     14.3G   0.04956  0.009492         0   0.05905        14       640: 100% 64/64 [02:04<00:00,  1.95s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  45% 5/11 [00:04<00:04,  1.29it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:06<00:03,  1.00it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.25s/it]\n",
            "                 all         347         369        0.77       0.623       0.685       0.327\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     18/31     14.3G   0.04552  0.009509         0   0.05502        62       640:  12% 8/64 [00:11<01:25,  1.53s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04377  0.009405         0   0.05318        56       640:  22% 14/64 [00:22<01:37,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04378  0.009324         0   0.05311        48       640:  23% 15/64 [00:24<01:30,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G    0.0443  0.009307         0   0.05361        53       640:  38% 24/64 [00:42<01:19,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04358  0.009236         0   0.05281        44       640:  39% 25/64 [00:44<01:24,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04376  0.009215         0   0.05298        53       640:  41% 26/64 [00:47<01:26,  2.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04386   0.00917         0   0.05303        47       640:  42% 27/64 [00:49<01:24,  2.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04449  0.009241         0   0.05373        44       640:  52% 33/64 [01:00<01:00,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04459  0.009191         0   0.05378        46       640:  53% 34/64 [01:04<01:11,  2.40s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04475  0.009108         0   0.05385        49       640:  58% 37/64 [01:08<00:45,  1.67s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04424  0.009151         0   0.05339        57       640:  61% 39/64 [01:12<00:44,  1.78s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G    0.0447  0.009161         0   0.05386        38       640:  69% 44/64 [01:22<00:36,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04471  0.009228         0   0.05394        82       640:  75% 48/64 [01:31<00:38,  2.40s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04414   0.00914         0   0.05328        59       640:  92% 59/64 [01:52<00:09,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04444  0.009151         0    0.0536        64       640:  94% 60/64 [01:56<00:09,  2.43s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04481  0.009162         0   0.05397        60       640:  95% 61/64 [01:57<00:06,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     18/31     14.3G   0.04479  0.009195         0   0.05398        10       640: 100% 64/64 [02:01<00:00,  1.90s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.22s/it]\n",
            "                 all         347         369       0.745       0.621       0.663       0.349\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "  0% 0/64 [00:00<?, ?it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.07003   0.01116         0   0.08119        59       640:   2% 1/64 [00:01<01:29,  1.41s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.05479   0.01031         0    0.0651        57       640:   5% 3/64 [00:03<01:19,  1.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04606  0.009738         0    0.0558        58       640:  11% 7/64 [00:08<01:10,  1.24s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04763  0.009587         0   0.05722        51       640:  16% 10/64 [00:15<01:50,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G    0.0475  0.009723         0   0.05722        55       640:  17% 11/64 [00:17<01:33,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04929  0.009911         0   0.05921        58       640:  22% 14/64 [00:25<01:57,  2.35s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G    0.0505  0.009937         0   0.06044        59       640:  23% 15/64 [00:26<01:41,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04779  0.009607         0    0.0574        45       640:  34% 22/64 [00:43<01:36,  2.29s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04642  0.009673         0   0.05609        56       640:  45% 29/64 [00:56<01:04,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04608  0.009652         0   0.05573        38       640:  48% 31/64 [01:00<01:05,  1.99s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04604  0.009629         0   0.05567        45       640:  56% 36/64 [01:13<01:04,  2.31s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04521  0.009679         0   0.05489        57       640:  64% 41/64 [01:22<00:43,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04517  0.009809         0   0.05498        68       640:  70% 45/64 [01:30<00:35,  1.87s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04508  0.009765         0   0.05484        47       640:  78% 50/64 [01:42<00:33,  2.36s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04515  0.009758         0   0.05491        53       640:  80% 51/64 [01:43<00:26,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04492  0.009722         0   0.05464        41       640:  86% 55/64 [01:52<00:19,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04501   0.00972         0   0.05473        62       640:  89% 57/64 [01:57<00:15,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04528  0.009742         0   0.05502        49       640:  92% 59/64 [02:04<00:13,  2.63s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     19/31     14.3G   0.04528  0.009815         0   0.05509        16       640: 100% 64/64 [02:14<00:00,  2.10s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  18% 2/11 [00:01<00:08,  1.11it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.30s/it]\n",
            "                 all         347         369       0.726       0.631       0.678       0.308\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     20/31     14.3G   0.03682  0.009696         0   0.04652        63       640:   8% 5/64 [00:07<01:36,  1.63s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04182  0.009534         0   0.05135        63       640:  12% 8/64 [00:12<01:27,  1.55s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04312  0.009384         0    0.0525        48       640:  16% 10/64 [00:17<01:45,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04261  0.009612         0   0.05222        47       640:  23% 15/64 [00:26<01:25,  1.75s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G    0.0431  0.009683         0   0.05279        52       640:  25% 16/64 [00:27<01:19,  1.65s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04632  0.009522         0   0.05584        65       640:  36% 23/64 [00:42<01:13,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04608  0.009505         0   0.05558        45       640:  38% 24/64 [00:45<01:34,  2.36s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04513  0.009487         0   0.05462        47       640:  39% 25/64 [00:47<01:28,  2.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04509  0.009532         0   0.05462        56       640:  42% 27/64 [00:51<01:17,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G    0.0454  0.009487         0   0.05488        57       640:  45% 29/64 [00:55<01:10,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04425  0.009604         0   0.05386        49       640:  55% 35/64 [01:07<00:51,  1.77s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04368  0.009555         0   0.05324        43       640:  56% 36/64 [01:09<00:53,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G    0.0438  0.009535         0   0.05333        47       640:  58% 37/64 [01:10<00:48,  1.79s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G    0.0435  0.009534         0   0.05304        44       640:  61% 39/64 [01:14<00:42,  1.70s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04413  0.009452         0   0.05358        49       640:  67% 43/64 [01:20<00:34,  1.63s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04432  0.009474         0    0.0538        61       640:  70% 45/64 [01:25<00:36,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04435  0.009341         0   0.05369        49       640:  89% 57/64 [01:52<00:15,  2.16s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04422  0.009325         0   0.05355        53       640:  97% 62/64 [02:05<00:05,  2.73s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     20/31     14.3G   0.04394   0.00929         0   0.05323        14       640: 100% 64/64 [02:06<00:00,  1.98s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.33s/it]\n",
            "                 all         347         369       0.765       0.664       0.706       0.362\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     21/31     14.3G   0.04959    0.0105         0    0.0601        69       640:   2% 1/64 [00:01<01:15,  1.20s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04166  0.009473         0   0.05114        42       640:   9% 6/64 [00:07<01:15,  1.31s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G    0.0434   0.01006         0   0.05346        59       640:  14% 9/64 [00:13<01:40,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04329   0.01019         0   0.05348        63       640:  17% 11/64 [00:16<01:30,  1.71s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04017  0.009996         0   0.05017        42       640:  20% 13/64 [00:21<01:35,  1.88s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04091  0.009898         0   0.05081        53       640:  22% 14/64 [00:24<02:00,  2.41s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04137   0.00969         0   0.05106        37       640:  23% 15/64 [00:26<01:41,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04157  0.009448         0   0.05102        43       640:  34% 22/64 [00:41<01:31,  2.19s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04161  0.009624         0   0.05124        70       640:  36% 23/64 [00:42<01:18,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G    0.0411  0.009618         0   0.05072        61       640:  39% 25/64 [00:47<01:20,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04129  0.009671         0   0.05096        65       640:  41% 26/64 [00:52<01:44,  2.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04167  0.009717         0   0.05139        71       640:  42% 27/64 [00:53<01:25,  2.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04206  0.009654         0   0.05171        55       640:  52% 33/64 [01:04<00:56,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04294  0.009767         0   0.05271        56       640:  72% 46/64 [01:28<00:34,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G    0.0429  0.009777         0   0.05267        71       640:  80% 51/64 [01:39<00:24,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     21/31     14.3G   0.04323  0.009563         0   0.05279         6       640: 100% 64/64 [02:06<00:00,  1.97s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:   9% 1/11 [00:00<00:06,  1.62it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.29s/it]\n",
            "                 all         347         369        0.88       0.634       0.754       0.378\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     22/31     14.3G    0.0462   0.01018         0   0.05638        44       640:  11% 7/64 [00:08<01:13,  1.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04908   0.01005         0   0.05913        53       640:  14% 9/64 [00:13<01:40,  1.83s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G    0.0487  0.009623         0   0.05832        53       640:  23% 15/64 [00:25<01:33,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04681  0.009642         0   0.05646        47       640:  27% 17/64 [00:30<01:39,  2.12s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04589  0.009467         0   0.05536        52       640:  31% 20/64 [00:37<01:48,  2.46s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04477  0.009395         0   0.05416        38       640:  33% 21/64 [00:39<01:33,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04469   0.00947         0   0.05416        64       640:  34% 22/64 [00:41<01:36,  2.31s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04287  0.009438         0   0.05231        46       640:  38% 24/64 [00:45<01:22,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04294  0.009285         0   0.05222        61       640:  47% 30/64 [00:57<01:18,  2.32s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04312  0.009251         0   0.05238        51       640:  48% 31/64 [00:59<01:06,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04224  0.009197         0   0.05144        45       640:  61% 39/64 [01:13<00:42,  1.69s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04215  0.009184         0   0.05133        40       640:  69% 44/64 [01:22<00:32,  1.63s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04245  0.009161         0   0.05161        54       640:  72% 46/64 [01:25<00:31,  1.73s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04252  0.009164         0   0.05169        51       640:  73% 47/64 [01:27<00:28,  1.65s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04316  0.009261         0   0.05242        47       640:  81% 52/64 [01:37<00:20,  1.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04317  0.009303         0   0.05247        61       640:  83% 53/64 [01:39<00:21,  1.91s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04359  0.009221         0   0.05281        52       640:  91% 58/64 [01:51<00:12,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04366  0.009208         0   0.05287        46       640:  92% 59/64 [01:54<00:11,  2.34s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     22/31     14.3G   0.04391  0.009202         0   0.05311        12       640: 100% 64/64 [02:03<00:00,  1.93s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.24s/it]\n",
            "                 all         347         369       0.723       0.664       0.715       0.307\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     23/31     14.3G   0.04442  0.009764         0   0.05418        77       640:  11% 7/64 [00:09<01:28,  1.55s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04223  0.009305         0   0.05153        44       640:  16% 10/64 [00:15<01:37,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04064   0.00895         0   0.04959        46       640:  23% 15/64 [00:24<01:29,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04162  0.008914         0   0.05054        62       640:  34% 22/64 [00:38<01:35,  2.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G    0.0407  0.008733         0   0.04943        40       640:  67% 43/64 [01:17<00:40,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04078  0.008801         0   0.04958        60       640:  72% 46/64 [01:24<00:36,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04084  0.008844         0   0.04969        54       640:  73% 47/64 [01:25<00:32,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04087  0.008893         0   0.04976        57       640:  75% 48/64 [01:28<00:33,  2.12s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04082  0.008941         0   0.04976        58       640:  81% 52/64 [01:37<00:26,  2.23s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04093  0.008985         0   0.04992        63       640:  83% 53/64 [01:39<00:21,  1.94s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04069  0.008985         0   0.04967        77       640:  92% 59/64 [01:52<00:09,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04041  0.008968         0   0.04938        52       640:  95% 61/64 [01:56<00:05,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     23/31     14.3G   0.04028  0.008955         0   0.04923         9       640: 100% 64/64 [02:03<00:00,  1.93s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:06<00:04,  1.19s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.20s/it]\n",
            "                 all         347         369       0.866       0.683       0.766       0.375\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     24/31     14.3G    0.0383  0.008476         0   0.04678        60       640:  11% 7/64 [00:09<01:26,  1.52s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G    0.0401  0.008564         0   0.04866        64       640:  23% 15/64 [00:27<01:34,  1.93s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G    0.0405  0.009046         0   0.04955        68       640:  34% 22/64 [00:41<01:27,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G   0.04059   0.00917         0   0.04976        62       640:  36% 23/64 [00:42<01:14,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G   0.04031  0.009211         0   0.04952        64       640:  48% 31/64 [00:59<00:59,  1.82s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G   0.04095   0.00922         0   0.05017        39       640:  56% 36/64 [01:09<00:54,  1.96s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G   0.04142  0.009249         0   0.05067        44       640:  64% 41/64 [01:21<00:46,  2.03s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G   0.04092  0.009183         0    0.0501        48       640:  70% 45/64 [01:30<00:40,  2.15s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     24/31     14.3G   0.04133  0.009225         0   0.05056        15       640: 100% 64/64 [02:11<00:00,  2.05s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  45% 5/11 [00:03<00:04,  1.40it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:06<00:04,  1.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.26s/it]\n",
            "                 all         347         369       0.867       0.707       0.799       0.419\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     25/31     14.3G   0.04345  0.008906         0   0.05236        59       640:   2% 1/64 [00:01<01:10,  1.11s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04407  0.009217         0   0.05329        56       640:  14% 9/64 [00:14<01:50,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04307   0.00922         0   0.05229        62       640:  22% 14/64 [00:24<01:42,  2.05s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04327  0.009128         0    0.0524        53       640:  25% 16/64 [00:28<01:29,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G    0.0418  0.009021         0   0.05082        40       640:  28% 18/64 [00:31<01:21,  1.76s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04283   0.00902         0   0.05184        50       640:  38% 24/64 [00:43<01:11,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04221  0.009036         0   0.05125        49       640:  42% 27/64 [00:50<01:10,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04068  0.008811         0   0.04949        45       640:  66% 42/64 [01:18<00:36,  1.68s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04023  0.008774         0     0.049        40       640:  67% 43/64 [01:20<00:37,  1.77s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04041  0.008773         0   0.04918        31       640:  70% 45/64 [01:24<00:34,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G    0.0405  0.008784         0   0.04928        49       640:  72% 46/64 [01:26<00:35,  1.99s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04099  0.008886         0   0.04988        49       640:  77% 49/64 [01:33<00:30,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04129  0.008917         0   0.05021        56       640:  84% 54/64 [01:44<00:23,  2.34s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     25/31     14.3G   0.04165  0.009021         0   0.05067        13       640: 100% 64/64 [02:05<00:00,  1.97s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:07<00:04,  1.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.21s/it]\n",
            "                 all         347         369       0.903       0.688        0.79       0.404\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     26/31     14.3G   0.04597  0.008704         0   0.05467        45       640:  11% 7/64 [00:09<01:15,  1.33s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G    0.0439   0.00842         0   0.05232        54       640:  19% 12/64 [00:19<01:45,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04349  0.008612         0    0.0521        57       640:  22% 14/64 [00:22<01:32,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04194  0.008702         0   0.05064        44       640:  25% 16/64 [00:26<01:31,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04213  0.008726         0   0.05086        57       640:  28% 18/64 [00:31<01:32,  2.02s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04214  0.008761         0    0.0509        53       640:  31% 20/64 [00:35<01:29,  2.04s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04224  0.008773         0   0.05101        46       640:  36% 23/64 [00:40<01:09,  1.69s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04135  0.008753         0   0.05011        54       640:  38% 24/64 [00:41<01:01,  1.55s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04135  0.008777         0   0.05013        55       640:  44% 28/64 [00:50<01:04,  1.79s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04107  0.008785         0   0.04986        52       640:  45% 29/64 [00:52<01:04,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04075  0.008828         0   0.04958        65       640:  59% 38/64 [01:09<00:43,  1.69s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04073  0.008661         0   0.04939        46       640:  72% 46/64 [01:26<00:37,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G    0.0407  0.008616         0   0.04931        49       640:  75% 48/64 [01:29<00:29,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G    0.0411  0.008628         0   0.04973        54       640:  78% 50/64 [01:33<00:25,  1.85s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04097  0.008646         0   0.04962        45       640:  86% 55/64 [01:46<00:22,  2.52s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G   0.04102  0.008643         0   0.04967        57       640:  88% 56/64 [01:47<00:16,  2.12s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     26/31     14.3G    0.0409  0.008542         0   0.04944        11       640: 100% 64/64 [02:04<00:00,  1.95s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.27s/it]\n",
            "                 all         347         369       0.843       0.732       0.819       0.437\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "  0% 0/64 [00:00<?, ?it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03896  0.008421         0   0.04738        52       640:   8% 5/64 [00:06<01:20,  1.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03808  0.008321         0    0.0464        48       640:  12% 8/64 [00:11<01:27,  1.57s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03809  0.008567         0   0.04665        58       640:  16% 10/64 [00:15<01:33,  1.72s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03702  0.008675         0   0.04569        46       640:  19% 12/64 [00:19<01:33,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03761  0.008761         0   0.04638        55       640:  22% 14/64 [00:23<01:30,  1.81s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03913  0.008689         0   0.04782        61       640:  31% 20/64 [00:35<01:23,  1.89s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03819  0.008628         0   0.04682        49       640:  36% 23/64 [00:40<01:08,  1.68s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03824  0.008623         0   0.04686        56       640:  38% 24/64 [00:41<01:07,  1.68s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G    0.0376  0.008638         0   0.04623        54       640:  39% 25/64 [00:43<01:07,  1.74s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03825  0.008628         0   0.04688        60       640:  45% 29/64 [00:52<01:04,  1.84s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03811  0.008555         0   0.04666        63       640:  56% 36/64 [01:07<01:04,  2.29s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03746  0.008556         0   0.04602        51       640:  61% 39/64 [01:12<00:46,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03755  0.008541         0   0.04609        56       640:  62% 40/64 [01:14<00:49,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03748  0.008571         0   0.04605        39       640:  70% 45/64 [01:22<00:30,  1.62s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03762  0.008591         0   0.04621        50       640:  75% 48/64 [01:29<00:34,  2.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03811   0.00857         0   0.04668        47       640:  81% 52/64 [01:36<00:22,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03877  0.008613         0   0.04738        37       640:  95% 61/64 [01:58<00:06,  2.13s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     27/31     14.3G   0.03858  0.008753         0   0.04733        17       640: 100% 64/64 [02:02<00:00,  1.92s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  18% 2/11 [00:01<00:07,  1.13it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.20s/it]\n",
            "                 all         347         369       0.856       0.713       0.791       0.406\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     28/31     14.3G    0.0361  0.007645         0   0.04374        44       640:   5% 3/64 [00:03<01:11,  1.17s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03508  0.008229         0   0.04331        55       640:  19% 12/64 [00:19<01:47,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03504   0.00823         0   0.04327        54       640:  20% 13/64 [00:21<01:38,  1.94s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03582  0.008295         0   0.04412        54       640:  23% 15/64 [00:26<01:43,  2.10s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03613  0.008483         0   0.04462        63       640:  27% 17/64 [00:29<01:24,  1.80s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03619  0.008429         0   0.04462        50       640:  28% 18/64 [00:31<01:34,  2.05s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03539  0.008345         0   0.04373        55       640:  34% 22/64 [00:40<01:24,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03606  0.008339         0    0.0444        61       640:  39% 25/64 [00:45<01:09,  1.79s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03599  0.008456         0   0.04444        51       640:  72% 46/64 [01:23<00:33,  1.86s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03625  0.008482         0   0.04473        52       640:  77% 49/64 [01:29<00:31,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03683  0.008447         0   0.04528        47       640:  83% 53/64 [01:37<00:21,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03674  0.008428         0   0.04517        54       640:  86% 55/64 [01:42<00:18,  2.07s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03676  0.008435         0   0.04519        39       640:  92% 59/64 [01:54<00:12,  2.53s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03698  0.008387         0   0.04537        44       640:  97% 62/64 [02:01<00:04,  2.31s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     28/31     14.3G   0.03706   0.00835         0   0.04541        14       640: 100% 64/64 [02:03<00:00,  1.92s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:   0% 0/11 [00:00<?, ?it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  36% 4/11 [00:03<00:05,  1.29it/s]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.28s/it]\n",
            "                 all         347         369       0.885       0.734       0.821       0.436\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     29/31     14.3G   0.03882   0.00905         0   0.04787        60       640:  23% 15/64 [00:27<01:37,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03954  0.008835         0   0.04837        57       640:  34% 22/64 [00:42<01:42,  2.43s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03852  0.008816         0   0.04733        45       640:  36% 23/64 [00:44<01:26,  2.10s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03811  0.008815         0   0.04692        57       640:  44% 28/64 [00:55<01:30,  2.52s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03765  0.008857         0   0.04651        60       640:  50% 32/64 [01:01<00:56,  1.75s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G    0.0377  0.008859         0   0.04656        62       640:  56% 36/64 [01:10<01:06,  2.37s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03764  0.008885         0   0.04652        47       640:  61% 39/64 [01:15<00:48,  1.92s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03791  0.009036         0   0.04695        71       640:  67% 43/64 [01:25<00:48,  2.29s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03798  0.009034         0   0.04701        56       640:  69% 44/64 [01:28<00:49,  2.45s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G    0.0377  0.009028         0   0.04673        60       640:  73% 47/64 [01:33<00:33,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03759  0.009032         0   0.04662        44       640:  77% 49/64 [01:37<00:30,  2.01s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03743  0.008965         0    0.0464        32       640:  83% 53/64 [01:43<00:18,  1.65s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03763  0.008967         0    0.0466        49       640:  89% 57/64 [01:53<00:14,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03784  0.008966         0   0.04681        53       640:  95% 61/64 [02:02<00:06,  2.06s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     29/31     14.3G   0.03775  0.009042         0   0.04679        21       640: 100% 64/64 [02:07<00:00,  1.99s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:14<00:00,  1.33s/it]\n",
            "                 all         347         369       0.876       0.729       0.822       0.434\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     30/31     14.3G    0.0415  0.008635         0   0.05013        56       640:  16% 10/64 [00:16<01:42,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03865  0.008201         0   0.04685        50       640:  23% 15/64 [00:26<01:38,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03754  0.008226         0   0.04576        52       640:  28% 18/64 [00:31<01:17,  1.68s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03594  0.008284         0   0.04422        41       640:  33% 21/64 [00:39<01:30,  2.11s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03629  0.008401         0   0.04469        67       640:  36% 23/64 [00:43<01:25,  2.08s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03669  0.008387         0   0.04507        54       640:  39% 25/64 [00:47<01:19,  2.03s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03702  0.008334         0   0.04535        56       640:  42% 27/64 [00:52<01:17,  2.10s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03715  0.008389         0   0.04554        55       640:  44% 28/64 [00:55<01:23,  2.33s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03657  0.008379         0   0.04494        60       640:  53% 34/64 [01:07<01:01,  2.05s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03577  0.008375         0   0.04414        47       640:  61% 39/64 [01:17<00:48,  1.95s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G    0.0359  0.008394         0    0.0443        41       640:  64% 41/64 [01:20<00:43,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03616  0.008306         0   0.04446        44       640:  70% 45/64 [01:26<00:30,  1.58s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03644  0.008356         0    0.0448        71       640:  78% 50/64 [01:39<00:30,  2.20s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03664   0.00836         0     0.045        54       640:  80% 51/64 [01:40<00:24,  1.90s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03667  0.008411         0   0.04508        62       640:  81% 52/64 [01:43<00:26,  2.22s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G    0.0366  0.008458         0   0.04506        79       640:  86% 55/64 [01:49<00:18,  2.09s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03596  0.008381         0   0.04434        48       640:  94% 60/64 [02:02<00:09,  2.39s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03613  0.008347         0   0.04448        55       640:  97% 62/64 [02:05<00:03,  2.00s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     30/31     14.3G   0.03621  0.008406         0   0.04462        20       640: 100% 64/64 [02:08<00:00,  2.00s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:13<00:00,  1.22s/it]\n",
            "                 all         347         369       0.923       0.718       0.839       0.429\n",
            "\n",
            "     Epoch   gpu_mem       box       obj       cls     total    labels  img_size\n",
            "     31/31     14.3G    0.0366   0.00857         0   0.04517        56       640:   9% 6/64 [00:07<01:14,  1.28s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03502   0.00857         0   0.04359        65       640:  17% 11/64 [00:20<02:09,  2.45s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03626  0.008508         0   0.04476        66       640:  22% 14/64 [00:25<01:38,  1.98s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03677  0.008379         0   0.04515        47       640:  31% 20/64 [00:36<01:10,  1.59s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03748  0.008417         0   0.04589        56       640:  36% 23/64 [00:43<01:27,  2.13s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03747  0.008375         0   0.04584        51       640:  38% 24/64 [00:45<01:15,  1.88s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03675  0.008492         0   0.04524        60       640:  41% 26/64 [00:49<01:11,  1.87s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03652  0.008339         0   0.04486        59       640:  47% 30/64 [00:58<01:16,  2.26s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03685  0.008423         0   0.04527        55       640:  55% 35/64 [01:08<01:06,  2.30s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03705  0.008483         0   0.04553        63       640:  62% 40/64 [01:15<00:37,  1.58s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G    0.0369  0.008427         0   0.04533        48       640:  64% 41/64 [01:17<00:38,  1.66s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03696  0.008453         0   0.04542        66       640:  72% 46/64 [01:28<00:33,  1.88s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03738  0.008441         0   0.04582        43       640:  86% 55/64 [01:49<00:22,  2.49s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03745   0.00847         0   0.04592        57       640:  89% 57/64 [01:55<00:17,  2.47s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03738  0.008427         0   0.04581        52       640:  92% 59/64 [01:59<00:10,  2.18s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "libpng warning: iCCP: known incorrect sRGB profile\n",
            "     31/31     14.3G   0.03726  0.008349         0   0.04561         6       640: 100% 64/64 [02:09<00:00,  2.02s/it]\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95:  64% 7/11 [00:09<00:04,  1.22s/it]libpng warning: iCCP: known incorrect sRGB profile\n",
            "               Class      Images      Labels           P           R      mAP@.5  mAP@.5:.95: 100% 11/11 [00:16<00:00,  1.54s/it]\n",
            "                 all         347         369       0.886       0.734       0.825       0.459\n",
            "32 epochs completed in 1.392 hours.\n",
            "\n",
            "Optimizer stripped from runs/train/yolov7x2/weights/last.pt, 142.1MB\n",
            "Optimizer stripped from runs/train/yolov7x2/weights/best.pt, 142.1MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "If you are using only CPU, try this:"
      ],
      "metadata": {
        "id": "0CE0CWiXP8ex"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# !python train.py --device cpu --batch-size 16 --data data/coco.yaml --img 640 640 --epochs 64 --weights yolov7x.pt --hyp data/hyp.scratch.p5.yaml --name yolov7x"
      ],
      "metadata": {
        "id": "Mw5WsRIzP_pU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Uplad Test Image and Video to Make Prediction"
      ],
      "metadata": {
        "id": "09ytXwsAiGbN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Video Testing:\n",
        "\n",
        "*   Select a video file that contains drone footage.\n",
        "*   Load the pre-trained YOLOv7 model and its corresponding weights.\n",
        "*   Utilize the OpenCV library to read and process each frame of the video.\n",
        "*   Pass each frame through the drone detection model for real-time inference.\n",
        "*   Draw bounding boxes around detected drones and display the annotated video output.\n",
        "\n",
        "\n",
        "Image Testing:\n",
        "\n",
        "*   Choose an image that includes a UAV.\n",
        "*   Load the pre-trained YOLOv7 model and its weights.\n",
        "*   Read and process the UAV image.\n",
        "*   Apply the trained model to the image and identify the presence of a drone.\n",
        "*   Visualize the image with a bounding box around the detected drone, if present.\n",
        "*   Analyze the model's performance by assessing the correct identification of the drone."
      ],
      "metadata": {
        "id": "JBOWYP4NwoOD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/"
      ],
      "metadata": {
        "id": "7dJQLwK9hlYq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2560cac7-6c68-4112-87ed-c5556e51aa2b"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!mkdir test"
      ],
      "metadata": {
        "id": "wzI4IJDyhplR"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/test"
      ],
      "metadata": {
        "id": "JaSlMix3h61W",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7a4897fa-9a13-4764-98f2-be669e4708ba"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/test\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import files\n",
        "uploaded = files.upload()"
      ],
      "metadata": {
        "id": "Fv43JlOwh94U",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        },
        "outputId": "71585725-7eac-42ba-ad61-9ffc8e860f65"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-b45679aa-d5fb-4969-a69b-7ad1c90a782a\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-b45679aa-d5fb-4969-a69b-7ad1c90a782a\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script>// Copyright 2017 Google LLC\n",
              "//\n",
              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
              "// you may not use this file except in compliance with the License.\n",
              "// You may obtain a copy of the License at\n",
              "//\n",
              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
              "//\n",
              "// Unless required by applicable law or agreed to in writing, software\n",
              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
              "// See the License for the specific language governing permissions and\n",
              "// limitations under the License.\n",
              "\n",
              "/**\n",
              " * @fileoverview Helpers for google.colab Python module.\n",
              " */\n",
              "(function(scope) {\n",
              "function span(text, styleAttributes = {}) {\n",
              "  const element = document.createElement('span');\n",
              "  element.textContent = text;\n",
              "  for (const key of Object.keys(styleAttributes)) {\n",
              "    element.style[key] = styleAttributes[key];\n",
              "  }\n",
              "  return element;\n",
              "}\n",
              "\n",
              "// Max number of bytes which will be uploaded at a time.\n",
              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
              "\n",
              "function _uploadFiles(inputId, outputId) {\n",
              "  const steps = uploadFilesStep(inputId, outputId);\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  // Cache steps on the outputElement to make it available for the next call\n",
              "  // to uploadFilesContinue from Python.\n",
              "  outputElement.steps = steps;\n",
              "\n",
              "  return _uploadFilesContinue(outputId);\n",
              "}\n",
              "\n",
              "// This is roughly an async generator (not supported in the browser yet),\n",
              "// where there are multiple asynchronous steps and the Python side is going\n",
              "// to poll for completion of each step.\n",
              "// This uses a Promise to block the python side on completion of each step,\n",
              "// then passes the result of the previous step as the input to the next step.\n",
              "function _uploadFilesContinue(outputId) {\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  const steps = outputElement.steps;\n",
              "\n",
              "  const next = steps.next(outputElement.lastPromiseValue);\n",
              "  return Promise.resolve(next.value.promise).then((value) => {\n",
              "    // Cache the last promise value to make it available to the next\n",
              "    // step of the generator.\n",
              "    outputElement.lastPromiseValue = value;\n",
              "    return next.value.response;\n",
              "  });\n",
              "}\n",
              "\n",
              "/**\n",
              " * Generator function which is called between each async step of the upload\n",
              " * process.\n",
              " * @param {string} inputId Element ID of the input file picker element.\n",
              " * @param {string} outputId Element ID of the output display.\n",
              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
              " */\n",
              "function* uploadFilesStep(inputId, outputId) {\n",
              "  const inputElement = document.getElementById(inputId);\n",
              "  inputElement.disabled = false;\n",
              "\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  outputElement.innerHTML = '';\n",
              "\n",
              "  const pickedPromise = new Promise((resolve) => {\n",
              "    inputElement.addEventListener('change', (e) => {\n",
              "      resolve(e.target.files);\n",
              "    });\n",
              "  });\n",
              "\n",
              "  const cancel = document.createElement('button');\n",
              "  inputElement.parentElement.appendChild(cancel);\n",
              "  cancel.textContent = 'Cancel upload';\n",
              "  const cancelPromise = new Promise((resolve) => {\n",
              "    cancel.onclick = () => {\n",
              "      resolve(null);\n",
              "    };\n",
              "  });\n",
              "\n",
              "  // Wait for the user to pick the files.\n",
              "  const files = yield {\n",
              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
              "    response: {\n",
              "      action: 'starting',\n",
              "    }\n",
              "  };\n",
              "\n",
              "  cancel.remove();\n",
              "\n",
              "  // Disable the input element since further picks are not allowed.\n",
              "  inputElement.disabled = true;\n",
              "\n",
              "  if (!files) {\n",
              "    return {\n",
              "      response: {\n",
              "        action: 'complete',\n",
              "      }\n",
              "    };\n",
              "  }\n",
              "\n",
              "  for (const file of files) {\n",
              "    const li = document.createElement('li');\n",
              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
              "    li.append(span(\n",
              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
              "        `last modified: ${\n",
              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
              "                                    'n/a'} - `));\n",
              "    const percent = span('0% done');\n",
              "    li.appendChild(percent);\n",
              "\n",
              "    outputElement.appendChild(li);\n",
              "\n",
              "    const fileDataPromise = new Promise((resolve) => {\n",
              "      const reader = new FileReader();\n",
              "      reader.onload = (e) => {\n",
              "        resolve(e.target.result);\n",
              "      };\n",
              "      reader.readAsArrayBuffer(file);\n",
              "    });\n",
              "    // Wait for the data to be ready.\n",
              "    let fileData = yield {\n",
              "      promise: fileDataPromise,\n",
              "      response: {\n",
              "        action: 'continue',\n",
              "      }\n",
              "    };\n",
              "\n",
              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
              "    let position = 0;\n",
              "    do {\n",
              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
              "      const chunk = new Uint8Array(fileData, position, length);\n",
              "      position += length;\n",
              "\n",
              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
              "      yield {\n",
              "        response: {\n",
              "          action: 'append',\n",
              "          file: file.name,\n",
              "          data: base64,\n",
              "        },\n",
              "      };\n",
              "\n",
              "      let percentDone = fileData.byteLength === 0 ?\n",
              "          100 :\n",
              "          Math.round((position / fileData.byteLength) * 100);\n",
              "      percent.textContent = `${percentDone}% done`;\n",
              "\n",
              "    } while (position < fileData.byteLength);\n",
              "  }\n",
              "\n",
              "  // All done.\n",
              "  yield {\n",
              "    response: {\n",
              "      action: 'complete',\n",
              "    }\n",
              "  };\n",
              "}\n",
              "\n",
              "scope.google = scope.google || {};\n",
              "scope.google.colab = scope.google.colab || {};\n",
              "scope.google.colab._files = {\n",
              "  _uploadFiles,\n",
              "  _uploadFilesContinue,\n",
              "};\n",
              "})(self);\n",
              "</script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving pexels-nino-souza-8459631-1920x1080-30fps.mp4 to pexels-nino-souza-8459631-1920x1080-30fps.mp4\n",
            "Saving pexels-petar-avramoski-6081353.jpg to pexels-petar-avramoski-6081353.jpg\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%cd /content/yolov7"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GB_TlOL3YqAU",
        "outputId": "c77078d6-7e2d-4acb-d449-cb5eccb45000"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/yolov7\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Upload your images into test folder."
      ],
      "metadata": {
        "id": "K11lx1fshuUk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!python detect.py --weights runs/train/yolov7x/weights/best.pt --conf 0.50 --img-size 640 --source /content/test/pexels-petar-avramoski-6081353.jpg"
      ],
      "metadata": {
        "id": "wV6OQ1qXhRiK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "bf675f20-8b87-4178-84f9-849751276991"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Namespace(weights=['runs/train/yolov7x/weights/best.pt'], source='/content/test/pexels-petar-avramoski-6081353.jpg', img_size=640, conf_thres=0.5, iou_thres=0.45, device='', view_img=False, save_txt=False, save_conf=False, nosave=False, classes=None, agnostic_nms=False, augment=False, update=False, project='runs/detect', name='exp', exist_ok=False, no_trace=False)\n",
            "YOLOR 🚀 v0.1-122-g3b41c2c torch 2.0.1+cu118 CUDA:0 (Tesla T4, 15101.8125MB)\n",
            "\n",
            "Fusing layers... \n",
            "Model Summary: 354 layers, 70780150 parameters, 0 gradients\n",
            " Convert model to Traced-model... \n",
            " traced_script_module saved! \n",
            " model is traced! \n",
            "\n",
            "/usr/local/lib/python3.10/dist-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3483.)\n",
            "  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]\n",
            "Done. (29.6ms) Inference, (0.6ms) NMS\n",
            " The image with the result is saved in: runs/detect/exp/pexels-petar-avramoski-6081353.jpg\n",
            "Done. (1.330s)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Image source: [@ernesto](https://www.pexels.com/@ernesto/)\n",
        "\n",
        "Image link: [pexels](https://www.pexels.com/photo/professional-drone-flying-against-mountains-6081353/)"
      ],
      "metadata": {
        "id": "N-66ix5n6SpQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In addition, you can upload video and make predictions."
      ],
      "metadata": {
        "id": "NsplmcWYiuFe"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!python detect.py --weights runs/train/yolov7x/weights/best.pt --conf 0.25 --img-size 640 --source /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4"
      ],
      "metadata": {
        "id": "CAK0wYFHi0w0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "56c075c0-a81e-43a9-8497-a86645c2e396"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Namespace(weights=['runs/train/yolov7x/weights/best.pt'], source='/content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4', img_size=640, conf_thres=0.25, iou_thres=0.45, device='', view_img=False, save_txt=False, save_conf=False, nosave=False, classes=None, agnostic_nms=False, augment=False, update=False, project='runs/detect', name='exp', exist_ok=False, no_trace=False)\n",
            "YOLOR 🚀 v0.1-122-g3b41c2c torch 2.0.1+cu118 CUDA:0 (Tesla T4, 15101.8125MB)\n",
            "\n",
            "Fusing layers... \n",
            "Model Summary: 354 layers, 70780150 parameters, 0 gradients\n",
            " Convert model to Traced-model... \n",
            " traced_script_module saved! \n",
            " model is traced! \n",
            "\n",
            "/usr/local/lib/python3.10/dist-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3483.)\n",
            "  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]\n",
            "video 1/1 (1/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (25.3ms) Inference, (0.5ms) NMS\n",
            "video 1/1 (2/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (25.3ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (3/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (25.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (4/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (25.4ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (5/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (6/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (21.6ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (7/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (21.5ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (8/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (9/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (10/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (11/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (12/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (13/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (14/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (15/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (16/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (17/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.8ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (18/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (19/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.7ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (20/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (21/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (22/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.2ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (23/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.2ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (24/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.2ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (25/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (26/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (27/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (28/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (29/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (30/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (31/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (32/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (33/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (34/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (35/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.9ms) NMS\n",
            "video 1/1 (36/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (37/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (38/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (39/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (40/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (41/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (42/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (43/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (16.9ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (44/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (45/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (46/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (47/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (48/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (49/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (50/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (51/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (52/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (53/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (54/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (55/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (56/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (57/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (58/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (59/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (60/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (61/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (2.2ms) NMS\n",
            "video 1/1 (62/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (63/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (64/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (65/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (66/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (67/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (68/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (69/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (70/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (71/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (72/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (73/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (74/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (75/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (76/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (77/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (78/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (79/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.9ms) NMS\n",
            "video 1/1 (80/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (81/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (82/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.9ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (83/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.9ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (84/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.8ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (85/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (86/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (87/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (88/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (89/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (90/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (91/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (92/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (93/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (94/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (95/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.7ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (96/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (97/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (98/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.9ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (99/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.9ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (100/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (101/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.6ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (102/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.6ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (103/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (104/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (105/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (106/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (107/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (108/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (109/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (110/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (111/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (112/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (113/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (114/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (115/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (116/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (117/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (118/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (119/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (120/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.5ms) NMS\n",
            "video 1/1 (121/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (122/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (123/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (124/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (125/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (126/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (127/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (128/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (129/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (130/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (131/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (132/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (133/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (134/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.0ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (135/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (136/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (137/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (138/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (139/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (140/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (141/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (142/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (143/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (144/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (145/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (146/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.2ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (147/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (148/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.0ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (149/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (150/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.6ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (151/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (152/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.6ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (153/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.6ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (154/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.7ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (155/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (156/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (157/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (158/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (16.5ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (159/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.5ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (160/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (161/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (162/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (163/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (164/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (165/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (166/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (167/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (168/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (169/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (170/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (171/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.0ms) NMS\n",
            "video 1/1 (172/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (173/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (174/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (175/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (176/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (177/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (178/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (179/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (180/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.7ms) Inference, (0.5ms) NMS\n",
            "video 1/1 (181/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (182/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (21.9ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (183/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (184/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (15.9ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (185/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.8ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (186/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (2.3ms) NMS\n",
            "video 1/1 (187/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.5ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (188/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (189/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (20.0ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (190/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (191/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (192/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (193/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (23.7ms) Inference, (3.5ms) NMS\n",
            "video 1/1 (194/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.9ms) NMS\n",
            "video 1/1 (195/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (196/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (197/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.0ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (198/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (17.8ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (199/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.0ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (200/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.9ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (201/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (21.9ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (202/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (23.1ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (203/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (19.3ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (204/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.7ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (205/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.7ms) Inference, (2.4ms) NMS\n",
            "video 1/1 (206/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (15.5ms) Inference, (0.5ms) NMS\n",
            "video 1/1 (207/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (19.0ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (208/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (17.0ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (209/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.6ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (210/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (20.9ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (211/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.7ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (212/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.1ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (213/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.0ms) Inference, (4.6ms) NMS\n",
            "video 1/1 (214/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.7ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (215/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.0ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (216/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (19.3ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (217/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.0ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (218/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (23.9ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (219/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (26.0ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (220/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.9ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (221/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (222/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (25.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (223/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.2ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (224/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (19.8ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (225/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (17.9ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (226/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.9ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (227/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.6ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (228/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (229/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.3ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (230/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.9ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (231/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (20.0ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (232/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (233/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.2ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (234/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.2ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (235/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.3ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (236/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (237/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.6ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (238/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (239/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (240/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (241/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (242/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.2ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (243/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (244/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (245/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (246/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (247/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (15.2ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (248/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (15.1ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (249/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.9ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (250/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (14.8ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (251/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (15.0ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (252/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.2ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (253/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (15.1ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (254/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.8ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (255/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.8ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (256/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.8ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (257/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (17.0ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (258/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (259/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (260/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (261/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (262/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (263/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (264/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (265/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (266/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (267/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (268/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (269/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (270/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (271/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.5ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (272/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (273/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (274/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (275/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (276/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (277/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (278/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (279/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (280/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (281/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.0ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (282/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (283/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (284/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (285/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (286/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (287/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (288/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (289/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (290/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (291/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (292/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (293/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (294/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (295/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (296/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (297/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.2ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (298/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (299/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (300/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (301/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.1ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (302/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (303/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (14.0ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (304/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (305/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (306/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (307/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (308/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (309/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (310/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (311/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (312/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (313/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (314/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (315/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (316/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (317/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (318/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.9ms) NMS\n",
            "video 1/1 (319/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (320/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (321/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (322/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (323/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (324/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (325/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (326/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (327/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 3 drones, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (328/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (329/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (330/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (331/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (332/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (333/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (334/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (335/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (336/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (337/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (338/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (339/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (340/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (341/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (342/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (343/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (344/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (345/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (346/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (347/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (348/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (349/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (350/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.0ms) NMS\n",
            "video 1/1 (351/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (352/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (353/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (354/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (355/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (356/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (357/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (358/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (359/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (360/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (361/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (362/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (363/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (364/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (365/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (366/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (367/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (368/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (369/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.9ms) Inference, (2.0ms) NMS\n",
            "video 1/1 (370/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (371/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (372/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (373/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (374/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (375/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (376/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (377/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (378/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (379/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (380/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (381/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (382/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (383/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (384/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (385/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (386/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (387/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (388/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (389/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (390/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (391/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (392/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (393/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.0ms) NMS\n",
            "video 1/1 (394/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (395/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (396/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (397/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.5ms) Inference, (2.8ms) NMS\n",
            "video 1/1 (398/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (399/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (400/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (401/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (402/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (403/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (404/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (405/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (406/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (407/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (408/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (409/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (410/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (411/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (412/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (413/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (414/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (415/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (416/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (417/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (418/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (419/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (420/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (421/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (422/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (423/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (424/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (425/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (426/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.6ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (427/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (428/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (429/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (430/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (431/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.8ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (432/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.0ms) NMS\n",
            "video 1/1 (433/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (434/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.9ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (435/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (436/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (437/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (438/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (439/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (440/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (441/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (442/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (443/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.7ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (444/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.7ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (445/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (446/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.7ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (447/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (448/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (449/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (450/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (451/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (452/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.6ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (453/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.6ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (454/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (455/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (456/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (457/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (458/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (459/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (460/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (461/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.0ms) NMS\n",
            "video 1/1 (462/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (463/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (464/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (465/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (466/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (467/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (468/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (469/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.0ms) NMS\n",
            "video 1/1 (470/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (471/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (472/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (473/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.7ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (474/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (475/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.5ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (476/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (477/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (478/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (479/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (480/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (481/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.2ms) Inference, (2.0ms) NMS\n",
            "video 1/1 (482/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (483/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (484/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (485/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (2.0ms) NMS\n",
            "video 1/1 (486/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.5ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (487/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (488/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (489/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.5ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (490/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (491/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (492/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (493/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (494/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (495/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.9ms) Inference, (1.2ms) NMS\n",
            "video 1/1 (496/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (497/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (498/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (499/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (500/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (1.9ms) NMS\n",
            "video 1/1 (501/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (502/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.1ms) NMS\n",
            "video 1/1 (503/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.9ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (504/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (505/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.9ms) Inference, (1.3ms) NMS\n",
            "video 1/1 (506/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (13.4ms) Inference, (2.3ms) NMS\n",
            "video 1/1 (507/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (508/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.9ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (509/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (510/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.6ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (511/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (512/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.4ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (513/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (17.1ms) Inference, (0.6ms) NMS\n",
            "video 1/1 (514/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (515/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (15.9ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (516/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (517/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.3ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (518/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (20.3ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (519/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (13.4ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (520/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.8ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (521/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.6ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (522/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (23.5ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (523/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.6ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (524/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.9ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (525/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.5ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (526/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (16.3ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (527/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (13.9ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (528/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (14.5ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (529/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (16.6ms) Inference, (0.3ms) NMS\n",
            "video 1/1 (530/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.1ms) Inference, (1.8ms) NMS\n",
            "video 1/1 (531/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (16.4ms) Inference, (1.4ms) NMS\n",
            "video 1/1 (532/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (19.7ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (533/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (534/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (16.8ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (535/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (16.7ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (536/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (19.7ms) Inference, (2.3ms) NMS\n",
            "video 1/1 (537/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (23.3ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (538/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.1ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (539/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (18.0ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (540/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.8ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (541/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 2 drones, Done. (17.4ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (542/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.2ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (543/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (21.1ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (544/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (19.4ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (545/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (17.6ms) Inference, (1.6ms) NMS\n",
            "video 1/1 (546/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.4ms) Inference, (4.5ms) NMS\n",
            "video 1/1 (547/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (18.3ms) Inference, (1.5ms) NMS\n",
            "video 1/1 (548/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: 1 drone, Done. (22.2ms) Inference, (1.7ms) NMS\n",
            "video 1/1 (549/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.6ms) Inference, (0.5ms) NMS\n",
            "video 1/1 (550/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.5ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (551/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.4ms) Inference, (0.4ms) NMS\n",
            "video 1/1 (552/552) /content/test/pexels-nino-souza-8459631-1920x1080-30fps.mp4: Done. (18.4ms) Inference, (0.4ms) NMS\n",
            "Done. (22.785s)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import moviepy.editor as mp\n",
        "from IPython.display import Image"
      ],
      "metadata": {
        "id": "ejhDDvJw3bQf"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "video_path = '/content/yolov7/runs/detect/exp3/pexels-nino-souza-8459631-1920x1080-30fps.mp4'  # Replace with the actual path to your video file\n",
        "gif_path = '/content/yolov7/runs/detect/exp3/pexels-nino-souza-8459631-1920x1080-30fps.gif'  # Replace with the desired path for the output GIF file\n",
        "\n",
        "clip = mp.VideoFileClip(video_path)\n",
        "clip.write_gif(gif_path)"
      ],
      "metadata": {
        "id": "jwk-qYZFNQXG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Image(gif_path)"
      ],
      "metadata": {
        "id": "G5WT78RrNaIX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Video source: [@ninosouza](https://www.pexels.com/@ninosouza/)\n",
        "\n",
        "Video link: [pexels](https://www.pexels.com/video/an-airborne-drone-machine-8459631/)"
      ],
      "metadata": {
        "id": "7bACoYsI3xRx"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Save the Results"
      ],
      "metadata": {
        "id": "BQ6sxcSCE2U0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "files.download('/content/yolov7/runs/train/yolov7x/F1_curve.png') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/PR_curve.png') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/confusion_matrix.png') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/hyp.yaml') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/opt.yaml')\n",
        "files.download('/content/yolov7/runs/train/yolov7x/results.png') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/results.txt') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/test_batch0_labels.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/test_batch0_pred.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/test_batch1_labels.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/test_batch1_pred.jpg')\n",
        "files.download('/content/yolov7/runs/train/yolov7x/test_batch2_labels.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/test_batch2_pred.jpg')\n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch0.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch1.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch2.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch3.jpg')\n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch4.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch5.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch6.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch7.jpg')\n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch8.jpg') \n",
        "files.download('/content/yolov7/runs/train/yolov7x/train_batch9.jpg') \n",
        "files.download('runs/detect/exp/pexels-petar-avramoski-6081353.jpg')\n",
        "files.download('/content/yolov7/data/coco.yaml')\n",
        "files.download('runs/train/yolov7x/weights/best.pt')"
      ],
      "metadata": {
        "id": "kOvj_lCGE5Xs",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "cca7f084-a9b2-4150-814d-89ff808ac6c2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_99c5e4af-a94e-441d-b9cc-7f722e679a34\", \"best.pt\", 142106919)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_5baa5b87-af2d-4921-92ae-7821a500b33d\", \"F1_curve.png\", 140423)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_b72a4749-1fd2-4602-a71f-c02bfb19a002\", \"PR_curve.png\", 112466)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_8a2afcea-0739-41a3-ac84-ea6f862f3741\", \"confusion_matrix.png\", 92405)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_8f0b61a3-7089-417b-bf0c-29dadf587726\", \"hyp.yaml\", 400)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_a202d144-5b46-408e-b43a-950b2d4afa63\", \"opt.yaml\", 670)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_c0283385-8c7e-42dc-8ea2-c3d40612e699\", \"results.png\", 276622)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_1ffabfad-fb7c-44b8-bb04-ccdaed5a730d\", \"results.txt\", 9664)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_dd1aa964-4f9f-4d78-9fdf-c495aad5725a\", \"test_batch0_labels.jpg\", 415433)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_ca62ba57-49b7-4bda-975b-2df7a7e190b7\", \"test_batch0_pred.jpg\", 419177)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_7fb3b976-c204-4ee6-b663-85e61982730d\", \"test_batch1_labels.jpg\", 448713)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_1fae5a44-8173-4a8c-80bb-ed866fd54528\", \"test_batch1_pred.jpg\", 453868)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_4fe00a36-e4db-4298-8ac7-d223d2c0a897\", \"test_batch2_labels.jpg\", 426470)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_79ca095f-697f-4751-b553-dacd0832fd24\", \"test_batch2_pred.jpg\", 434925)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_a56aa71a-6df6-4874-a2d5-468ef39f8e3c\", \"train_batch0.jpg\", 320543)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_2ce34ffc-24eb-48dd-89bc-b5fce139ef90\", \"train_batch1.jpg\", 370759)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_170e2c16-6447-434c-8561-24225d498595\", \"train_batch2.jpg\", 412515)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_5dd07a41-f30d-4bd4-b3f4-ce002aa3de57\", \"train_batch3.jpg\", 376350)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_6242e4f5-2b9b-47f5-aae1-dfb69a1e73e2\", \"train_batch4.jpg\", 400050)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_6e142b15-db0a-4c02-9ba7-f35af5c9a07d\", \"train_batch5.jpg\", 367233)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_4cb52b26-eefc-4e8f-af9b-663061b41961\", \"train_batch6.jpg\", 351567)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_19abbc45-1791-4455-baa9-f64cd8cb0744\", \"train_batch7.jpg\", 355022)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_4f529a41-022a-4e41-964c-093757aedf43\", \"train_batch8.jpg\", 286646)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_c24aec99-8662-4adb-b605-c06ece49f1c1\", \"train_batch9.jpg\", 366954)"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_c5d25fc9-f048-4887-b8d9-369fa0c5aded\", \"agri_0_14.jpeg\", 159806)"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Results and Discussion\n",
        "\n",
        "We present comprehensive results of our drone detection model's performance on both the training and testing datasets. The evaluation metrics include precision, recall, and F1-score, which are standard measures to assess the model's detection accuracy. Additionally, we analyze the model's performance across various environmental conditions and discuss its strengths and limitations.\n",
        "\n",
        "## Conclusion\n",
        "Our research addresses the critical need for reliable drone detection systems by proposing a comprehensive dataset and a state-of-the-art detection model using the YOLOv7 architecture. The availability of our curated dataset and the promising performance of our model offer valuable contributions to the field of drone detection. The outcomes of this study can pave the way for enhanced security measures and privacy protection in areas where drones pose potential risks.\n",
        "\n",
        "## Future Work\n",
        "Moving forward, further research can be conducted to explore the integration of our drone detection model into real-time surveillance systems and intelligent decision-making frameworks. Additionally, the dataset can be expanded to include a wider range of challenging scenarios and variations in drone types, enabling the development of more robust and generalized drone detection models."
      ],
      "metadata": {
        "id": "9H4yhgdkK3Zb"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Contact Me\n",
        "<p>If you have something to say to me please contact me:</p>\n",
        "\n",
        "<ul>\n",
        "  <li>Twitter: <a href=\"https://twitter.com/Doguilmak\">Doguilmak</a></li>\n",
        "  <li>Mail address: doguilmak@gmail.com</li>\n",
        "</ul>"
      ],
      "metadata": {
        "id": "-1-U5fDNFlI_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from datetime import datetime\n",
        "print(f\"Changes have been made to the project on {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\")"
      ],
      "metadata": {
        "id": "QZfvhOQKFlmq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4977651b-b87c-41d6-dd36-cd08f566776a"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Changes have been made to the project on 2023-05-28 16:47:57\n"
          ]
        }
      ]
    }
  ]
}