Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Modelzoo] Add K600, K700 baselines #259

Merged
merged 6 commits into from
Oct 17, 2020
Merged

[Modelzoo] Add K600, K700 baselines #259

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
24d7dc6
add training configs for K600, K700
Oct 16, 2020
fd5e93f
remove deprecated args
Oct 16, 2020
a03a4f7
update README file
Oct 16, 2020
585a24e
Merge branch 'master' into kinetics_ext_baselines
Oct 16, 2020
64676f9
update changelog (also add missing changelog for #256)
Oct 16, 2020
1474d5f
fix tsn k700 config
Oct 16, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 14 additions & 0 deletions configs/recognition/slowonly/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,20 @@ In data benchmark, we compare two different data preprocessing methods: (1) Resi
| [slowonly_r101_8x8x1_196e_kinetics400_rgb](/configs/recognition/slowonly/slowonly_r101_8x8x1_196e_kinetics400_rgb.py) | x | ResNet101 | None | :x: | 76.5 | 92.7 | x | x | [ckpt](https://download.openmmlab.com/mmaction/recognition/slowonly/omni/slowonly_r101_without_omni_8x8x1_kinetics400_rgb_20200926-0c730aef.pth) | x | x |
| x | x | ResNet101 | None | :heavy_check_mark: | 80.4 | 94.4 | x | x | [ckpt](https://download.openmmlab.com/mmaction/recognition/slowonly/omni/slowonly_r101_omni_8x8x1_kinetics400_rgb_20200926-b5dbb701.pth) | x | x |

### Kinetics-600

| config | resolution | gpus | backbone | pretrain | top1 acc | top5 acc | inference_time(video/s) | gpu_mem(M) | ckpt | log | json |
| :----------------------------------------------------------- | :------------: | :--: | :------: | :------: | :------: | :------: | :---------------------: | :--------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
| [slowonly_r50_video_8x8x1_256e_kinetics600_rgb](/configs/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics600_rgb.py) | short-side 256 | 8x4 | ResNet50 | None | 77.5 | 93.7 | 2.3 (80x3 frames) | 8478 | [ckpt](https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics600_rgb/slowonly_r50_video_8x8x1_256e_kinetics600_rgb_20201015-81e5153e.pth) | [log](https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics600_rgb/slowonly_r50_video_8x8x1_256e_kinetics600_rgb_20201015.log) | [json](https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics600_rgb/slowonly_r50_video_8x8x1_256e_kinetics600_rgb_20201015.json) |

### Kinetics-700

| config | resolution | gpus | backbone | pretrain | top1 acc | top5 acc | inference_time(video/s) | gpu_mem(M) | ckpt | log | json |
| :----------------------------------------------------------- | :------------: | :--: | :------: | :------: | :------: | :------: | :---------------------: | :--------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
| [slowonly_r50_video_8x8x1_256e_kinetics700_rgb](/configs/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics700_rgb.py) | short-side 256 | 8x4 | ResNet50 | None | 65.0 | 86.1 | 2.3 (80x3 frames) | 8478 | [ckpt](https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics700_rgb/slowonly_r50_video_8x8x1_256e_kinetics700_rgb_20201015-9250f662.pth) | [log](https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics700_rgb/slowonly_r50_video_8x8x1_256e_kinetics700_rgb_20201015.log) | [json](https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics700_rgb/slowonly_r50_video_8x8x1_256e_kinetics700_rgb_20201015.json) |



Notes:

1. The **gpus** indicates the number of gpu we used to get the checkpoint. It is noteworthy that the configs we provide are used for 8 gpus as default.
Expand Down
114 changes: 114 additions & 0 deletions configs/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics600_rgb.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,114 @@
model = dict(
type='Recognizer3D',
backbone=dict(
type='ResNet3dSlowOnly',
depth=50,
pretrained=None,
lateral=False,
conv1_kernel=(1, 7, 7),
conv1_stride_t=1,
pool1_stride_t=1,
inflate=(0, 0, 1, 1),
norm_eval=False),
cls_head=dict(
type='I3DHead',
in_channels=2048,
num_classes=600,
spatial_type='avg',
dropout_ratio=0.5))
train_cfg = None
test_cfg = dict(average_clips='prob')
dataset_type = 'VideoDataset'
data_root = 'data/kinetics600/videos_train'
data_root_val = 'data/kinetics600/videos_val'
ann_file_train = 'data/kinetics600/kinetics600_train_list_videos.txt'
ann_file_val = 'data/kinetics600/kinetics600_val_list_videos.txt'
ann_file_test = 'data/kinetics600/kinetics600_val_list_videos.txt'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
train_pipeline = [
dict(type='DecordInit'),
dict(type='SampleFrames', clip_len=8, frame_interval=8, num_clips=1),
dict(type='DecordDecode'),
dict(type='RandomResizedCrop'),
dict(type='Resize', scale=(224, 224), keep_ratio=False),
dict(type='Flip', flip_ratio=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCTHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs', 'label'])
]
val_pipeline = [
dict(type='DecordInit'),
dict(
type='SampleFrames',
clip_len=8,
frame_interval=8,
num_clips=1,
test_mode=True),
dict(type='DecordDecode'),
dict(type='CenterCrop', crop_size=224),
dict(type='Flip', flip_ratio=0),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCTHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs'])
]
test_pipeline = [
dict(type='DecordInit'),
dict(
type='SampleFrames',
clip_len=8,
frame_interval=8,
num_clips=10,
test_mode=True),
dict(type='DecordDecode'),
dict(type='ThreeCrop', crop_size=256),
dict(type='Flip', flip_ratio=0),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCTHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs'])
]
data = dict(
videos_per_gpu=12,
workers_per_gpu=4,
train=dict(
type=dataset_type,
ann_file=ann_file_train,
data_prefix=data_root,
pipeline=train_pipeline),
val=dict(
type=dataset_type,
ann_file=ann_file_val,
data_prefix=data_root_val,
pipeline=val_pipeline),
test=dict(
type=dataset_type,
ann_file=ann_file_test,
data_prefix=data_root_val,
pipeline=test_pipeline))
# optimizer
optimizer = dict(
type='SGD', lr=0.15, momentum=0.9,
weight_decay=0.0001) # this lr is used for 8 gpus
optimizer_config = dict(grad_clip=dict(max_norm=40, norm_type=2))
# learning policy
lr_config = dict(policy='CosineAnnealing', min_lr=0)
total_epochs = 256
checkpoint_config = dict(interval=4)
workflow = [('train', 1)]
evaluation = dict(
interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'], topk=(1, 5))
log_config = dict(
interval=20,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook'),
])
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/slowonly_r50_video_8x8x1_256e_kinetics600_rgb'
load_from = None
resume_from = None
find_unused_parameters = False
114 changes: 114 additions & 0 deletions configs/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics700_rgb.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,114 @@
model = dict(
type='Recognizer3D',
backbone=dict(
type='ResNet3dSlowOnly',
depth=50,
pretrained=None,
lateral=False,
conv1_kernel=(1, 7, 7),
conv1_stride_t=1,
pool1_stride_t=1,
inflate=(0, 0, 1, 1),
norm_eval=False),
cls_head=dict(
type='I3DHead',
in_channels=2048,
num_classes=700,
spatial_type='avg',
dropout_ratio=0.5))
train_cfg = None
test_cfg = dict(average_clips='prob')
dataset_type = 'VideoDataset'
data_root = 'data/kinetics700/videos_train'
data_root_val = 'data/kinetics700/videos_val'
ann_file_train = 'data/kinetics700/kinetics700_train_list_videos.txt'
ann_file_val = 'data/kinetics700/kinetics700_val_list_videos.txt'
ann_file_test = 'data/kinetics700/kinetics700_val_list_videos.txt'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
train_pipeline = [
dict(type='DecordInit'),
dict(type='SampleFrames', clip_len=8, frame_interval=8, num_clips=1),
dict(type='DecordDecode'),
dict(type='RandomResizedCrop'),
dict(type='Resize', scale=(224, 224), keep_ratio=False),
dict(type='Flip', flip_ratio=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCTHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs', 'label'])
]
val_pipeline = [
dict(type='DecordInit'),
dict(
type='SampleFrames',
clip_len=8,
frame_interval=8,
num_clips=1,
test_mode=True),
dict(type='DecordDecode'),
dict(type='CenterCrop', crop_size=224),
dict(type='Flip', flip_ratio=0),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCTHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs'])
]
test_pipeline = [
dict(type='DecordInit'),
dict(
type='SampleFrames',
clip_len=8,
frame_interval=8,
num_clips=10,
test_mode=True),
dict(type='DecordDecode'),
dict(type='ThreeCrop', crop_size=256),
dict(type='Flip', flip_ratio=0),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCTHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs'])
]
data = dict(
videos_per_gpu=12,
workers_per_gpu=4,
train=dict(
type=dataset_type,
ann_file=ann_file_train,
data_prefix=data_root,
pipeline=train_pipeline),
val=dict(
type=dataset_type,
ann_file=ann_file_val,
data_prefix=data_root_val,
pipeline=val_pipeline),
test=dict(
type=dataset_type,
ann_file=ann_file_test,
data_prefix=data_root_val,
pipeline=test_pipeline))
# optimizer
optimizer = dict(
type='SGD', lr=0.15, momentum=0.9,
weight_decay=0.0001) # this lr is used for 8 gpus
optimizer_config = dict(grad_clip=dict(max_norm=40, norm_type=2))
# learning policy
lr_config = dict(policy='CosineAnnealing', min_lr=0)
total_epochs = 256
checkpoint_config = dict(interval=4)
workflow = [('train', 1)]
evaluation = dict(
interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'], topk=(1, 5))
log_config = dict(
interval=20,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook'),
])
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/slowonly_r50_video_8x8x1_256e_kinetics700_rgb'
load_from = None
resume_from = None
find_unused_parameters = False
12 changes: 12 additions & 0 deletions configs/recognition/tsn/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,6 +74,18 @@ In data benchmark, we compare:

[1] We obtain the pre-trained model from [torch-hub](https://pytorch.org/hub/facebookresearch_semi-supervised-ImageNet1K-models_resnext/), the pretrain model we used is `resnet50_swsl`

### Kinetics-600

| config | resolution | gpus | backbone | pretrain | top1 acc | top5 acc | inference_time(video/s) | gpu_mem(M) | ckpt | log | json |
| :----------------------------------------------------------- | :------------: | :--: | :------: | :------: | :------: | :------: | :---------------------: | :--------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
| [tsn_r50_video_1x1x8_100e_kinetics600_rgb](/configs/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb.py) | short-side 256 | 8x2 | ResNet50 | ImageNet | 74.8 | 92.3 | 11.1 (25x3 frames) | 8344 | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb/tsn_r50_video_1x1x8_100e_kinetics600_rgb_20201015-4db3c461.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb/tsn_r50_video_1x1x8_100e_kinetics600_rgb_20201015.log) | [json](https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb/tsn_r50_video_1x1x8_100e_kinetics600_rgb_20201015.json) |

### Kinetics-700

| config | resolution | gpus | backbone | pretrain | top1 acc | top5 acc | inference_time(video/s) | gpu_mem(M) | ckpt | log | json |
| :----------------------------------------------------------- | :------------: | :--: | :------: | :------: | :------: | :------: | :---------------------: | :--------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
| [tsn_r50_video_1x1x8_100e_kinetics700_rgb](/configs/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb.py) | short-side 256 | 8x2 | ResNet50 | ImageNet | 61.7 | 83.6 | 11.1 (25x3 frames) | 8344 | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb/tsn_r50_video_1x1x8_100e_kinetics700_rgb_20201015-e381a6c7.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb/tsn_r50_video_1x1x8_100e_kinetics700_rgb_20201015.log) | [json](https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb/tsn_r50_video_1x1x8_100e_kinetics700_rgb_20201015.json) |

### Something-Something V1

|config|resolution | gpus| backbone |pretrain| top1 acc| top5 acc | reference top1 acc | reference top5 acc | gpu_mem(M) | ckpt | log| json|
Expand Down
116 changes: 116 additions & 0 deletions configs/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,116 @@
# model settings
model = dict(
type='Recognizer2D',
backbone=dict(
type='ResNet',
pretrained='torchvision://resnet50',
depth=50,
norm_eval=False),
cls_head=dict(
type='TSNHead',
num_classes=600,
in_channels=2048,
spatial_type='avg',
consensus=dict(type='AvgConsensus', dim=1),
dropout_ratio=0.4,
init_std=0.01))
# model training and testing settings
train_cfg = None
test_cfg = dict(average_clips=None)
# dataset settings
dataset_type = 'VideoDataset'
data_root = 'data/kinetics600/videos_train'
data_root_val = 'data/kinetics600/videos_val'
ann_file_train = 'data/kinetics600/kinetics600_train_list_videos.txt'
ann_file_val = 'data/kinetics600/kinetics600_val_list_videos.txt'
ann_file_test = 'data/kinetics600/kinetics600_val_list_videos.txt'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
train_pipeline = [
dict(type='DecordInit'),
dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
dict(type='DecordDecode'),
dict(type='RandomResizedCrop'),
dict(type='Resize', scale=(224, 224), keep_ratio=False),
dict(type='Flip', flip_ratio=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs', 'label'])
]
val_pipeline = [
dict(type='DecordInit'),
dict(
type='SampleFrames',
clip_len=1,
frame_interval=1,
num_clips=8,
test_mode=True),
dict(type='DecordDecode'),
dict(type='Resize', scale=(-1, 256)),
dict(type='CenterCrop', crop_size=256),
dict(type='Flip', flip_ratio=0),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs'])
]
test_pipeline = [
dict(type='DecordInit'),
dict(
type='SampleFrames',
clip_len=1,
frame_interval=1,
num_clips=25,
test_mode=True),
dict(type='DecordDecode'),
dict(type='Resize', scale=(-1, 256)),
dict(type='ThreeCrop', crop_size=256),
dict(type='Flip', flip_ratio=0),
dict(type='Normalize', **img_norm_cfg),
dict(type='FormatShape', input_format='NCHW'),
dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
dict(type='ToTensor', keys=['imgs'])
]
data = dict(
videos_per_gpu=12,
workers_per_gpu=4,
train=dict(
type=dataset_type,
ann_file=ann_file_train,
data_prefix=data_root,
pipeline=train_pipeline),
val=dict(
type=dataset_type,
ann_file=ann_file_val,
data_prefix=data_root_val,
pipeline=val_pipeline),
test=dict(
type=dataset_type,
ann_file=ann_file_test,
data_prefix=data_root_val,
pipeline=test_pipeline))
# optimizer
optimizer = dict(
type='SGD', lr=0.00375, momentum=0.9,
weight_decay=0.0001) # this lr is used for 8 gpus
optimizer_config = dict(grad_clip=dict(max_norm=40, norm_type=2))
# learning policy
lr_config = dict(policy='step', step=[40, 80])
total_epochs = 100
checkpoint_config = dict(interval=5)
evaluation = dict(
interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'], topk=(1, 5))
log_config = dict(
interval=20,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook'),
])
# runtime settings
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/tsn_r50_1x1x3_100e_kinetics600_rgb/'
load_from = None
resume_from = None
workflow = [('train', 1)]
Loading