Model_VGG16_as_features_extractor.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Model  VGG16 as features extractor",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/serg-kas/A-G-I-demo/blob/master/Model_VGG16_as_features_extractor.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8-F71M7l0j9Y"
      },
      "source": [
        "# Analog Gauge Inspection\n",
        "\n",
        "**Определение показаний аналогового стрелочного прибора.**\n",
        "\n",
        "![Voltmeter-Original.jpg](data:image/jpeg;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/serg-kas/A-G-I-demo\n",
        "%cd A-G-I-demo"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z8zjVFLCGJ8n",
        "outputId": "65890ec1-d802-4f41-975c-3b60468121de"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'A-G-I-demo'...\n",
            "remote: Enumerating objects: 172, done.\u001b[K\n",
            "remote: Counting objects: 100% (172/172), done.\u001b[K\n",
            "remote: Compressing objects: 100% (143/143), done.\u001b[K\n",
            "remote: Total 172 (delta 93), reused 71 (delta 28), pack-reused 0\u001b[K\n",
            "Receiving objects: 100% (172/172), 1.73 MiB | 7.14 MiB/s, done.\n",
            "Resolving deltas: 100% (93/93), done.\n",
            "/content/A-G-I-demo\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "psPMK8Iyr3Bf"
      },
      "source": [
        "# Импорт библиотек\n",
        "import numpy as np\n",
        "import cv2 as cv\n",
        "import os\n",
        "import random\n",
        "#\n",
        "import utils as u \n",
        "\n",
        "from tensorflow.keras.models import Model \n",
        "from tensorflow.keras.applications.vgg16 import VGG16\n",
        "\n",
        "from sklearn.metrics.pairwise import cosine_similarity\n",
        "\n",
        "from google.colab.patches import cv2_imshow\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "    # Подготовим изображение прибора (резайз и пр.)\n",
        "    u.gauge_img_preparing()\n",
        "    assert 'Voltmeter-Blank.jpg' in os.listdir('.')\n",
        "\n",
        "    # Загружаем изображение прибора без стрелки\n",
        "    img = cv.imread('Voltmeter-Blank.jpg')\n",
        "\n",
        "    # Подготавливаем данные положения стрелки прибора\n",
        "    # L - длинна стрелки\n",
        "    # angle_0, angle_1  - углы крайних положений стрелки\n",
        "    L, angle_0, angle_1 = u.gauge_needle_preparing(img)\n",
        "\n",
        "    # Получаем случайное показание прибора\n",
        "    img_example, r = u.get_random_measurement(img, L, angle_0, angle_1)\n",
        "    print('Диапазон движение стрелки: от {:.2f} до {:.2f}'.format(angle_0, angle_1))\n",
        "    print('Случайное показание прибора: {}'.format(int(300 * r)))\n",
        "    print('Размер картинки: {}'.format(img_example.shape))\n",
        "    \n",
        "    cv2_imshow(img_example)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 244
        },
        "id": "4WgRj8yqPKQ7",
        "outputId": "d89c4196-0b45-4ff9-a160-a00b0fd8a88f"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Диапазон движение стрелки: от 0.77 до 2.37\n",
            "Случайное показание прибора: 217\n",
            "Размер картинки: (175, 511, 3)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=511x175 at 0x7FAAEDDE6710>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf8AAACvCAIAAAB8akCqAAARYElEQVR4nO2d67KrKBBGYeq8/yszP7KTGAXs5mYDa9WuM5MEEQU+2gZaH0Lw3rs3ITjnggMAgKXxp8/hq/yMAQAAy/Lf6bM/DwcAALAgJ/UPjgEAAGADjuqPqwcAYBf+/Xn6D/O+j5UFAABG8e/9Px/5DwH9BwBYnf+c+/H0I/0AADvwH5O8AAAbElnzAwAAy/Pv+AG3D4BZjg/p9FSo5999EgB4Du/ddU/++zNrNKCcn1ZFUwKwQ1T3Y9BxQconqlsI4RzpAQAsIJZ+eTKAL957PD+g5rhMDKuzGwpN995j/4MW1B+k+Nji4Nd3SE9vTnc4WhcnvLsG7aKitieET6sgzg/c4/2N3HgvkSMo5KrZeRX3r+EhUiWemtqcY7v52v4YBRDFf/65S+hpRO0ottIFEwbee54C4BLfH+CM3FbEqnwa+VwxdQX4/XfkJBEZK/DqJjjajFet8TgQHyYq6iH2PXW1O6j/XmRmbiW+gFOSwwTSIS9cCv1J1GNmqA6po2L5/P33/APeorXA87MRN52/aD7wKgi4FLri7x7IMl+eFw5dM/+bFPbxZwhmjNcC9d8FWcdt0rtRiF4kVvK8f1JnV5AD+r8OqP8i+LfhJvQJpDNS9278AQPILLoVGv7RXA/5M7e/Hfj9V+HQJ0/TedfeejNzW71xFOd/Q/LLeF411cMj82kDsRZC/a4Atv8KnDt/KiKkcy42cws2ebv4kyZ/cpBOfJ8aI/ITCSG4y+ogzP8VQP2nJzo9p8qh3nVzyQF1qCK9Wde5vO6/EuQyvic2XSw5DiYD9Z+fDk/9KU8CLt8BZHTf3em+kqjXCKXfBdR/blo5fMWagvx3JLqa849bg19PNDvM/H1A/ScmI/31g8Irc9bzDCMWktM5d6/7V6d9Mn+AA6j/rNxZ/QWdXST0bPjpRXJO9rYuZTUiWgrMYL8RrPjcjtRyvUjYBjCB/x0Abp7HhE9riV0CqoLB3GD7r0DUN5BWcq3GE8vBFDcx+qO/VNaY0LkEc4H6T0/SFPx22eJI8S/Xfy5raIV2iiW5Mij+Za3+Hz/wiLAGqP+shDfHrxIpz99opSAiTL/mJ+7jJoQDqsN+P1/irsZOFM1IcVKYH/z+6xAN4p4gEdtd5/vPrE4U5wEJfqNx5ColUu+HYB3JpUSRMyqLCJOD+q/Md4K3eko3hCBc7cMi0eZI3rwgj9ipdzHxYLcmeH7W5q/fXvtrXBxOX54jAt13e6T/EeR3PVNBQuuAGl4G1H8pVOJ7a8pf88q7pJH+B2kwNlfGBofZwPOzCzHXzU8sZ3k3R+Vtko/2LKm1U3DvfGxwmB3Uf3HuQrHn4sp0KRB0pvrdDO8Wg9W/Onh+ViMTbBlBhxPxJpHYSkD7WQzUfy+EHZh+vg80iW1B/dfnspAnO3PrmgcSBuvc1jhNYknw+2/KdYYw9gI/2IUQQvQFwuj+wuAUXpPfyVzq9o9Vb8t57n6dK4P2fFoLtv/6FCjBjxm4rJQkwl3MBkvyoQzUf01qFPusJouIJAD8wKwv/BCN/zC8FFAFgzVIQP0BAHYE9Yd7cCybBac/FIP6ww+p94SNLQWUs+wkPbQG9QcA2BHUH85gPM4CXh+oAfUHEfiXTUKlQDms94cV+A5OjZ5cfmNg5GNkf5K1OHEdPLeBHNQfIsjf4tsVuQQfjxkuw/79H6QXZgL1BymnFz+BCu++jwlNbmPD4blklIX5we8PD+DfPF2QWhQX0P9irQ0qYBzUHxJ0swGP6jK91BzK3+p++bFjI3K/Lag/xIlq2a1Q+PefmNmlp3H5hVKMZEM9qD+oyImOf70PNvFW2CK+Y9A2cvfwdeL33wfUH0YjV5dfIZpY/k/vXhEepZLhMtXmGWJnUH9IEhUUoVogKj+c36BpFAz/rUD9QUnGWjxqh9ioxPrUgsEOTUD9oRkKu/HHxkTL/qhQ9fxu5PgaIkaRzfmoP098ECHu/KkWDeOtrcxN3/bMKidMJq1c4nH77Maf+lPve1K6Qkd0wMSW5cRFl4LhD3h+NudvgX4fLWiS52HR58Czfk/fZYVOObGaSp4aiYcMqD/coXNBlK07lJ3c2Hqjn2ILr7vDU7Ywy3zVhBAab9UA86D+8Eeq2yc2/bYViWpdf8Z3eXDTZxJ5YbLjp76Xk5gBVm7ThslB/eFD+56fNepbCtxvXpNK2P2eANWYy3wv5PlTfx73FkMYKUza7cudPzSstpzvp7AG715QM8dmNGgLtv/qaAb21GBRFvENbMJUMLxA/denjfmvzliZV7vMGpjDqxnA+ctRXOwyL2YAx7u9tkDzssN02qCR+2/i3BvBQpCovvA1kxLRlw4MrtNKp6phSaW5KWdOLBNe7LUpf+pPpU+E5DXiJ8VU6X9UpKJCncpWpuqK8cSmKmlL1WRHbk0BKsHkvyVxi2y237f6P/AqbCjn08LkdRbX9L9cyt/hnstWxcLNb3zXL1s8+qBErfFi4XQn8t5bvK6P359RfT7yil3c2kaaeOHNsDPugdztI86xm0Bv8kRh8MmJWd+pafdS8XZOcyggHIgmuGpHqH/q6vfqZt22YXOy2AlrAwCzvr3w3n2adT/dzLvsfl06WefP5WXrwiKnJnUZKrqT2hEmz0Fszv/I1l3NHhLftwFrglhDOiCu0WvE9u+C//zz+tgx8Fg78/83rTwp2CKh6a10Nq/olSdZz2g4XZCp0Q7bfwyi2VFh0DCVna4w/yVzv8LVPNCazrLYzPCXxKuI57yi9L9RrZYeB7b/IKRjvsxjWhxNQZ40WoT4UlAwT775yeuwoeFvygrujNErRf27ED7/HFA1gaYdqeR1u7ewWMc4TfagVRj+2s0Q0zel6UJk4/npRcRHottVceMski/Sl3trJHku0EvXpuEetHPK7K/FW5EX4rO/PZPGUPfB9u/ItR/ePIBrEqtyPuYtHzNgDYqltlii71YQ6dLb4Rs6t/zONC1QHaj/aFRbtFrt51LYd2y/2osbg/6bTr4k9J4pDf/65xVr3QrPT19ivhRVdARF4lw8NYA3r5c4vv+/S/6ZXyPb1lZptHnHacywezgoELZ/d9T+n17m/yJ9DFREG1sIf3/iAxsa/qeSTNks01Frzz9cH6bfnqNMUKBBz0ao/xi0TVy1oFPs05d0eliaMrVVeQ4zvy462ZtEa/lpk1XyUf9uoQgOC6E2q/ov1x6hWn+tTJxO+f4DkKN5K+Sak71yMh3V5zpyKkREXzrG949e6uu79Wr9lqtPMO+mPwfoR7VBjzGfjK4wn8IblItk503ccJ/46ZBJcOf66t7pe3l+8s2Ol8Op0UwVAERp3k40hr9usteiPlyeyI9l9D5ZZh/ze0RX1g3uyF3UX+7bMljF/eg6/QtwwnLv0vqILBALqX1c/x8f3hKz7ibMtfYrPjODfPSdomssUxQ+pWr9P7/xoW5itAmLCtCEGsP/Nm9hzoPRbLB/SX/qFxP0XfNzerpJXLefc/NHHEHjuD4BpJMaDg8Lc/GU6lyb7Hw+nwOS23hIcw5ierfKtrhcJTRWf0mUmIgDZCX5v1+hHzkim96MqQDzMPj1nLmfbc08NyCzGT64709FVTBUCZ/Z66t3gMyFOkDbWpcP5ujRuiR51kz2Gu8Rzbc0j3/bwdv2H258F++DsMllhrbltRjvBgBC2vp8aiOuDUQr/WNo7ffX6NTOA0DB+h+Cr4EQI/2oJkitTbks4/5aZIf4HxoUrLH6a2MUx0SwaYHGUhmiZ6IGDRNhxGhoGJNqpZ7i/VUpznZkTO8bDAHd4/zoC2ixXv37T3ucfgBQnwPgiE1l7LfOx8jYVkZio8D719s7UzcEvNW/3Q28VMYSCigOVHS9/E4lAjBOOJBJVuPzmV/6z/wsGeocEm5MjE+t/8euYkpudKX/B6AhczWwJX0+0SLfS3+M91Dapk67qP/aHm39AKB4+pmqq4I5putoXWeGzfD127/CAUXXg+al/5gghDaLDHvZ/tEBILup1XRFVt9r0SY44zcBpsN4izp1I1VphbsNnloSGitePBzQKXF+YGhLR89P4vrjiWewWHQx17ruAABYDIHPR5HYGehxQsm+lf7K/DN0j/Nz+S4xFD9dVbcUeLMYAGAkczaw4Ny9bXuwmmd6RM6b7adfa/ZFl9E90sM1qoH7ljuE8JL9Odx5l2u5iejwOmiOBxtYDrP96IOwgAUyN2xd0Oc8mZP0CIlRk/jDiDU/6ZIlHWGDkfsHtea8+Q4IizCn4X/PeR+UoEP93oqu0v/ZCFR1++uDApUx6K3u8lVKzxoskluqHwAYAQBKUVrxl3VB7UsUPVFDK3bYstdB6u9iq5RiaR4XStmN1Q8ALOmBkazR2LQyN2xJaMMnrZqFT8WHvBin/i+iIhh6rmrSI5jOjRzTpywAAlZ1+xzRWsRTSP8rv6fO/lh8/0fOm+I6nXsbbz+E8DtffX8IwBjWa4iq9aCS9O+jvCq9S4tvk3te4POpOe8z6i9n2Nx9wXqe94IlxSEAIEcs4u5oQWulX05X6S84deV5R3t+VJwuuPemvet0rnZFP8B4dnD73IH0l2Ba/SN0HgEuN5QBYHf8IcCrn+EN1Ls1yHodr9k53PZux7ZG9ZJ+Z9/zE8N7r7t41erfIhfQXv3NGm2dofmcce6Zol7HH5f+k+CMXEFu2vbPXKd6wH8bb7L1nERogDnYuXEuIP3arNoamqbV37UdAD6HddjSBaZoUl3RTIxb/ts+hvaQ/oy52Pw+SzJsflLr6u9aDACX4xkAlqdJZVmvcdrkiz7SnwxC02mITe156rdXVL3c9SmyDV1U/EsOJUfZv1EbEm8b1VXVb7atFf0mPBajSPqjrHCDP1c3ge3/InvXSwK0Cd0DlycAwTEwlnjbqKuq6MHGu77x4j1FK+k3FI6gEdOov2sS7fqUg3T96Fp1DhLMj/O4fSQ0lP5mZTLDTOrvqgeAULii/+9QhoG5aCuPxvu/8eI9AtKfZzL1dy2eAMoGgM8fGCTRKgrlH7N6AbSRf3aTfjej+jvBAKDfoNs7igRAS2iuKpD+KFOqv3vVys08sCAH7TFgmGhHbVWnxlXAePGe4y94fD7RntLv5lV/F3fiH7k35xkA4Ir9NmC/hBYQumq3lX43tfq/qJ0HZgBYiV7m//pCsCEjt/LaZHr1dx0GAJiUThVpvIHQgAs4vRLgyD73cwX1dwwA0Aj7T372S2if8VEcbLKI+rt0lIwXsgGARf3T03zu17gcGC+eQbJbefe6meuo/4u6AYAV/VuDVb08mzv6T6ym/u5uAKCHr0/c/Jccab1x0Hx7sKf0uyXV37UICQfzkqj7kko3rgvGizcFO9/DNdXfNQkJB2AJGm09J1nYWfrdwurvvhO5cehLC1Mw90t72IRw4OmyPMzK6u++4TkB1BhXB+PFA/ssrv7ubyUP/WRDFHO/9g1/+yWE6Vhf/V8wAOxGosKlGmq8wRgvHkzBLurvmPABADiwkfo7NvRuhnDu175TxX4JYUb+PV2A0WDxwy3GnwuNFw9mYS/bH8AR0QHAOYf6w9rcvu/XvlPFfglhUlB/gB+M+1WMFw8mAvWHxcnM/do3qjH8oR+oP2wM2gobg/rD+sjD/hj3qxgvHswF6g9gFNw+0BXUH+CDacsawx/agvrDFkikE3WFrUD9ASyC2wd6g/rDLuTNf+N+FePFgxlB/QHMgeEPA0D9AQB2BPWHjUj5T4z7VYwXDyYF9QewBW4fGMN28f1hc6azo6crMMwCtj8AwI6g/gCGwO0Dw8DzA4vQXzcf8MHg9oF+YPvD9PhBJnP3U2D4w0hQf5gfNBNAD+oPYBTcPtAV1B9ASF8txu0Dg2HWF6YnhOQrGoP7iPZkdjSGP/QG9YcVQCsBtOD5AXge3D4wHtQfwBw8ysAAUH8AgB1B/QEeBrcPPALqD2AL3D4whv8BlVOBAoPqJp0AAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Подготовим выборку из 60 картинок**"
      ],
      "metadata": {
        "id": "3goio06PD0D_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Подготовим набор картинок\n",
        "img = cv.imread('Voltmeter-Blank.jpg')\n",
        "angles = np.linspace(angle_0, angle_1, num=60)\n",
        "\n",
        "images = []\n",
        "for i in range(angles.shape[0]):\n",
        "  angle = angles[i]\n",
        "  # print(angle) \n",
        "  curr_img = img.copy()\n",
        "  curr_img = u.get_angle_measurement(curr_img, L, angle)\n",
        "  images.append(curr_img)\n",
        "\n",
        "# Примеры\n",
        "fig, axs = plt.subplots(1, 5, figsize=(18, 5))                      \n",
        "for i in range(5):                                                 \n",
        "    plot_img = random.choice(images)     \n",
        "    axs[i].imshow(plot_img)\n",
        "plt.show() "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 111
        },
        "id": "fkdXQfPzEYeX",
        "outputId": "2cbdb10a-bed8-4906-ea26-09ac2d27c0bd"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1296x360 with 5 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Модель на базе VGG16 как экстрактора фич**"
      ],
      "metadata": {
        "id": "BNN4s4nTvddI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# скачиваем модель\n",
        "# base_model = VGG16(include_top=False, weights='imagenet', input_shape=(224, 224, 3))\n",
        "base_model = VGG16(weights='imagenet')\n"
      ],
      "metadata": {
        "id": "cv5vA1KIvil4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5063e61d-d6a0-403c-fba9-a476a5fa45ff"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels.h5\n",
            "553467904/553467096 [==============================] - 3s 0us/step\n",
            "553476096/553467096 [==============================] - 3s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Экстрактор фич - это предобученная модель без последнего полносвязного слоя\n",
        "feat_extractor = Model(inputs=base_model.input, outputs=base_model.get_layer(\"fc2\").output)\n",
        "feat_extractor.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3u3ObrP6mdtu",
        "outputId": "62955385-86a4-488b-fc18-ff7bd615a42e"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
            "                                                                 \n",
            " block1_conv1 (Conv2D)       (None, 224, 224, 64)      1792      \n",
            "                                                                 \n",
            " block1_conv2 (Conv2D)       (None, 224, 224, 64)      36928     \n",
            "                                                                 \n",
            " block1_pool (MaxPooling2D)  (None, 112, 112, 64)      0         \n",
            "                                                                 \n",
            " block2_conv1 (Conv2D)       (None, 112, 112, 128)     73856     \n",
            "                                                                 \n",
            " block2_conv2 (Conv2D)       (None, 112, 112, 128)     147584    \n",
            "                                                                 \n",
            " block2_pool (MaxPooling2D)  (None, 56, 56, 128)       0         \n",
            "                                                                 \n",
            " block3_conv1 (Conv2D)       (None, 56, 56, 256)       295168    \n",
            "                                                                 \n",
            " block3_conv2 (Conv2D)       (None, 56, 56, 256)       590080    \n",
            "                                                                 \n",
            " block3_conv3 (Conv2D)       (None, 56, 56, 256)       590080    \n",
            "                                                                 \n",
            " block3_pool (MaxPooling2D)  (None, 28, 28, 256)       0         \n",
            "                                                                 \n",
            " block4_conv1 (Conv2D)       (None, 28, 28, 512)       1180160   \n",
            "                                                                 \n",
            " block4_conv2 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
            "                                                                 \n",
            " block4_conv3 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
            "                                                                 \n",
            " block4_pool (MaxPooling2D)  (None, 14, 14, 512)       0         \n",
            "                                                                 \n",
            " block5_conv1 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
            "                                                                 \n",
            " block5_conv2 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
            "                                                                 \n",
            " block5_conv3 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
            "                                                                 \n",
            " block5_pool (MaxPooling2D)  (None, 7, 7, 512)         0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 25088)             0         \n",
            "                                                                 \n",
            " fc1 (Dense)                 (None, 4096)              102764544 \n",
            "                                                                 \n",
            " fc2 (Dense)                 (None, 4096)              16781312  \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 134,260,544\n",
            "Trainable params: 134,260,544\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Получаем фичи картинок\n",
        "features = []\n",
        "for curr_img in images:\n",
        "  img_to_pred = curr_img.copy()\n",
        "  img_to_pred = cv.resize(img_to_pred, (224, 224), cv.INTER_LINEAR)\n",
        "  img_to_pred = img_to_pred / 255. # preprocess_input ?\n",
        "  img_to_pred = np.expand_dims(img_to_pred, axis = 0)\n",
        "  pred = feat_extractor.predict(img_to_pred)[0]\n",
        "  features.append(pred)\n"
      ],
      "metadata": {
        "id": "9wDaA-4HnXmJ"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Сохраним массивы\n",
        "images_np = np.array(images)\n",
        "feat_np = np.array(features)\n",
        "#\n",
        "np.save('angles', angles)\n",
        "np.save('images', images_np)\n",
        "np.save('feat', feat_np)"
      ],
      "metadata": {
        "id": "10QHijARsXsP"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Посмотрим предикт**"
      ],
      "metadata": {
        "id": "fBJdqhE46DKY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# \n",
        "img = cv.imread('Voltmeter-Blank.jpg')\n",
        "L, angle_0, angle_1 = u.gauge_needle_preparing(img)\n",
        "\n",
        "N = 5\n",
        "fig, axs = plt.subplots(2, N, figsize=(18, 5))\n",
        "for i in range(N):\n",
        "  curr_img, curr_r = u.get_random_measurement(img.copy(), L, angle_0, angle_1)\n",
        "  angle_r = angle_0 + curr_r * (angle_1 - angle_0)\n",
        "\n",
        "  # получаем предикт\n",
        "  res_img, res_r = u.get_pred(feat_extractor, feat_np, angles, curr_img.copy(), L, angle_0, angle_1)\n",
        "  \n",
        "  # пересчитаем радианы в вольты\n",
        "  curr_V = int(300 * curr_r)\n",
        "  res_V = int(300 * res_r)\n",
        "\n",
        "  axs[0,i].imshow(curr_img)\n",
        "  axs[0,i].set_title(curr_V)\n",
        "  axs[1,i].imshow(res_img)\n",
        "  axs[1,i].set_title(res_V)\n",
        "\n",
        "plt.show() \n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 271
        },
        "id": "4h282b2BWy-7",
        "outputId": "b4a4acb2-106f-4719-9951-0c7d73fcc52d"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1296x360 with 10 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Выполним 100 предиктов и посчитаем среднюю точность. \n",
        "\n",
        "Желательно получить точность не хуже половины деления шкалы, что составляет в данном случае 5В"
      ],
      "metadata": {
        "id": "TC56E0im5eHh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# \n",
        "img = cv.imread('Voltmeter-Blank.jpg')\n",
        "L, angle_0, angle_1 = u.gauge_needle_preparing(img)\n",
        "\n",
        "acc_list = []  # список для подсчета точности\n",
        "\n",
        "N = 100\n",
        "for i in range(N):\n",
        "  curr_img, curr_r = u.get_random_measurement(img.copy(), L, angle_0, angle_1)\n",
        "  angle_r = angle_0 + curr_r * (angle_1 - angle_0)\n",
        "  \n",
        "  # получаем предикт\n",
        "  res_img, res_r = u.get_pred(feat_extractor, feat_np, angles, curr_img.copy(), L, angle_0, angle_1)\n",
        "\n",
        "  # пересчитаем радианы в вольты\n",
        "  curr_V = int(300 * curr_r)\n",
        "  res_V = int(300 * res_r)\n",
        "\n",
        "  acc_list.append(abs(curr_V - res_V))\n",
        "\n",
        "acc_np = np.array(acc_list)\n",
        "print('Средняя погрешность, В: {}'.format(acc_np.mean()))\n",
        "print('Максимальная погрешность, В: {}'.format(acc_np.max()))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ANVLJWlY5KaZ",
        "outputId": "33afb259-821c-426f-8d72-373182884a5d"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Средняя погрешность, В: 1.48\n",
            "Максимальная погрешность, В: 4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "При выборке из 60 картинок получилась средняя погрешность 1,48В, максимальная 4В\n",
        "\n",
        "При выборке из 30 картинок получилась средняя погрешность 2,84В, максимальная 7В\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "oAr1KDMP9rw_"
      }
    }
  ]
}