diff --git a/.gitignore b/.gitignore index 020cf31531..722a50e523 100644 --- a/.gitignore +++ b/.gitignore @@ -128,5 +128,7 @@ work_dirs/ !tests/data/**/*.pkl.json !tests/data/**/*.log.json !tests/data/**/*.pth + +# avoid soft links created by MIM mmaction/configs/* mmaction/tools/* diff --git a/configs/detection/acrn/metafile.yml b/configs/detection/acrn/metafile.yml new file mode 100644 index 0000000000..fe41d6db10 --- /dev/null +++ b/configs/detection/acrn/metafile.yml @@ -0,0 +1,46 @@ +Collections: +- Name: ACRN + README: configs/detection/acrn/README.md +Models: +- Config: configs/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb.py + In Collection: ACRN + Metadata: + Architecture: ResNet50 + Batch Size: 6 + Epochs: 10 + Input: 32x2 + Modality: RGB + Parameters: 92232057 + Pretrained: Kinetics-400 + Training Data: AVA v2.1 + Training Resources: 8 GPUs + Name: slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb + Results: + - Dataset: AVA v2.1 + Metrics: + - mAP: 27.1 + Task: Spatial Temporal Action Detection + Training Json Log: https://download.openmmlab.com/mmaction/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb.json + Training Log: https://download.openmmlab.com/mmaction/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb.log + Weights: https://download.openmmlab.com/mmaction/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava_rgb-49b07bf2.pth +- Config: configs/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.py + In Collection: ACRN + Metadata: + Architecture: ResNet50 + Batch Size: 6 + Epochs: 10 + Input: 32x2 + Modality: RGB + Parameters: 92232057 + Pretrained: Kinetics-400 + Training Data: AVA v2.2 + Training Resources: 8 GPUs + Name: slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb + Results: + - Dataset: AVA v2.2 + Metrics: + - mAP: 27.8 + Task: Spatial Temporal Action Detection + Training Json Log: https://download.openmmlab.com/mmaction/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.json + Training Log: https://download.openmmlab.com/mmaction/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.log + Weights: https://download.openmmlab.com/mmaction/detection/acrn/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_acrn_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb-2be32625.pth diff --git a/configs/detection/ava/metafile.yml b/configs/detection/ava/metafile.yml index 520076cf6f..abb89ce202 100644 --- a/configs/detection/ava/metafile.yml +++ b/configs/detection/ava/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: AVA +- Name: AVA README: configs/detection/ava/README.md Models: - Config: configs/detection/ava/slowonly_kinetics_pretrained_r50_4x16x1_20e_ava_rgb.py In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 20 Input: 4x16 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 16 Training Data: AVA v2.1 - gpus: 8 + Training Resources: 8 GPUs Modality: RGB Name: slowonly_kinetics_pretrained_r50_4x16x1_20e_ava_rgb Results: @@ -29,13 +27,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 20 Input: 4x16 Pretrained: OmniSource Resolution: short-side 256 - Training BatchSize / GPU: 16 Training Data: AVA v2.1 - gpus: 8 + Training Resources: 8 GPUs Modality: RGB Name: slowonly_omnisource_pretrained_r50_4x16x1_20e_ava_rgb Results: @@ -50,13 +48,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 10 Input: 4x16 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 12 Training Data: AVA v2.1 - gpus: 8 + Training Resources: 8 GPUs Modality: RGB Name: slowonly_nl_kinetics_pretrained_r50_4x16x1_10e_ava_rgb Results: @@ -71,13 +69,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 10 Input: 8x8 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 6 Training Data: AVA v2.1 - gpus: 16 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_nl_kinetics_pretrained_r50_8x8x1_10e_ava_rgb Results: @@ -92,13 +90,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet101 + Batch Size: 6 Epochs: 20 Input: 8x8 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 6 Training Data: AVA v2.1 - gpus: 16 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_kinetics_pretrained_r101_8x8x1_20e_ava_rgb Results: @@ -113,13 +111,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet101 + Batch Size: 6 Epochs: 20 Input: 8x8 Pretrained: OmniSource Resolution: short-side 256 - Training BatchSize / GPU: 6 Training Data: AVA v2.1 - gpus: 16 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_omnisource_pretrained_r101_8x8x1_20e_ava_rgb Results: @@ -134,13 +132,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 9 Epochs: 20 Input: 32x2 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 9 Training Data: AVA v2.1 - gpus: 16 + Training Resources: 16 GPUs Modality: RGB Name: slowfast_kinetics_pretrained_r50_4x16x1_20e_ava_rgb Results: @@ -155,13 +153,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 9 Epochs: 20 Input: 32x2 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 9 Training Data: AVA v2.1 - gpus: 16 + Training Resources: 16 GPUs Modality: RGB Name: slowfast_context_kinetics_pretrained_r50_4x16x1_20e_ava_rgb Results: @@ -176,13 +174,13 @@ Models: In Collection: AVA Metadata: Architecture: ResNet50 + Batch Size: 5 Epochs: 20 Input: 32x2 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 5 Training Data: AVA v2.1 - gpus: 16 + Training Resources: 16 GPUs Modality: RGB Name: slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb Results: @@ -193,3 +191,66 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb/slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb_20201217.json Training Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb/slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb_20201217.log Weights: https://download.openmmlab.com/mmaction/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb/slowfast_kinetics_pretrained_r50_8x8x1_20e_ava_rgb_20201217-ae225e97.pth +- Config: configs/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.py + In Collection: AVA + Metadata: + Architecture: ResNet50 + Batch Size: 6 + Epochs: 10 + Input: 32x2 + Pretrained: Kinetics-400 + Resolution: short-side 256 + Training Data: AVA v2.2 + Training Resources: 8 GPUs + Modality: RGB + Name: slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb + Results: + - Dataset: AVA v2.2 + Metrics: + mAP: 26.1 + Task: Spatial Temporal Action Detection + Training Json Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.json + Training Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.log + Weights: https://download.openmmlab.com/mmaction/detection/ava/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb-b987b516.pth +- Config: configs/detection/ava/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.py + In Collection: AVA + Metadata: + Architecture: ResNet50 + Batch Size: 6 + Epochs: 10 + Input: 32x2 + Pretrained: Kinetics-400 + Resolution: short-side 256 + Training Data: AVA v2.2 + Training Resources: 8 GPUs + Modality: RGB + Name: slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb + Results: + - Dataset: AVA v2.2 + Metrics: + mAP: 26.8 + Task: Spatial Temporal Action Detection + Training Json Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.json + Training Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.log + Weights: https://download.openmmlab.com/mmaction/detection/ava/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_temporal_max_focal_alpha3_gamma1_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb-345618cd.pth +- Config: configs/detection/ava/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.py + In Collection: AVA + Metadata: + Architecture: ResNet50 + Batch Size: 6 + Epochs: 10 + Input: 32x2 + Pretrained: Kinetics-400 + Resolution: short-side 256 + Training Data: AVA v2.2 + Training Resources: 8 GPUs + Modality: RGB + Name: slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb + Results: + - Dataset: AVA v2.2 + Metrics: + mAP: 26.4 + Task: Spatial Temporal Action Detection + Training Json Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.json + Training Log: https://download.openmmlab.com/mmaction/detection/ava/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb.log + Weights: https://download.openmmlab.com/mmaction/detection/ava/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb/slowfast_temporal_max_kinetics_pretrained_r50_8x8x1_cosine_10e_ava22_rgb-874e0845.pth diff --git a/configs/detection/lfb/metafile.yml b/configs/detection/lfb/metafile.yml index 6e9a09f18f..e9b15fc94b 100644 --- a/configs/detection/lfb/metafile.yml +++ b/configs/detection/lfb/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: LFB +- Name: LFB README: configs/detection/lfb/README.md Models: - Config: configs/detection/lfb/lfb_nl_kinetics_pretrained_slowonly_r50_4x16x1_20e_ava_rgb.py In Collection: LFB Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 20 Input: 4x16 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 12 Training Data: AVA v2.1 - gpus: 8 + Training Resources: 8 GPUs Modality: RGB Name: lfb_nl_kinetics_pretrained_slowonly_r50_4x16x1_20e_ava_rgb.py Results: @@ -29,13 +27,13 @@ Models: In Collection: LFB Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 20 Input: 4x16 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 12 Training Data: AVA v2.1 - gpus: 8 + Training Resources: 8 GPUs Modality: RGB Name: lfb_avg_kinetics_pretrained_slowonly_r50_4x16x1_20e_ava_rgb.py Results: @@ -50,13 +48,13 @@ Models: In Collection: LFB Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 20 Input: 4x16 Pretrained: Kinetics-400 Resolution: short-side 256 - Training BatchSize / GPU: 12 Training Data: AVA v2.1 - gpus: 8 + Training Resources: 8 GPUs Modality: RGB Name: lfb_max_kinetics_pretrained_slowonly_r50_4x16x1_20e_ava_rgb.py Results: diff --git a/configs/localization/bmn/metafile.yml b/configs/localization/bmn/metafile.yml index ac60106a46..f2d52ca714 100644 --- a/configs/localization/bmn/metafile.yml +++ b/configs/localization/bmn/metafile.yml @@ -1,17 +1,15 @@ Collections: -- Metadata: - Training Data: null - Name: BMN +- Name: BMN README: configs/localization/bmn/README.md Models: - Config: configs/localization/bmn/bmn_400x100_2x8_9e_activitynet_feature.py In Collection: BMN Metadata: + Batch Size: 8 Epochs: 9 - Training BatchSize / GPU: 8 Training Data: ActivityNet v1.3 + Training Resources: 2 GPUs feature: cuhk_mean_100 - gpus: 2 Name: bmn_400x100_9e_2x8_activitynet_feature (cuhk_mean_100) Results: - Dataset: ActivityNet v1.3 @@ -26,16 +24,14 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_9e_activitynet_feature/bmn_400x100_9e_activitynet_feature.log.json Training Log: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_9e_activitynet_feature/bmn_400x100_9e_activitynet_feature.log Weights: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_9e_activitynet_feature/bmn_400x100_9e_activitynet_feature_20200619-42a3b111.pth - gpu_mem(M): '5420' - iter time(s): '3.27' - Config: configs/localization/bmn/bmn_400x100_2x8_9e_activitynet_feature.py In Collection: BMN Metadata: + Batch Size: 8 Epochs: 9 - Training BatchSize / GPU: 8 Training Data: ActivityNet v1.3 + Training Resources: 2 GPUs feature: mmaction_video - gpus: 2 Name: bmn_400x100_9e_2x8_activitynet_feature (mmaction_video) Results: - Dataset: ActivityNet v1.3 @@ -50,16 +46,14 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_2x8_9e_mmaction_video/bmn_400x100_2x8_9e_mmaction_video_20200809.json Training Log: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_2x8_9e_mmaction_video/bmn_400x100_2x8_9e_mmaction_video_20200809.log Weights: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_2x8_9e_mmaction_video/bmn_400x100_2x8_9e_mmaction_video_20200809-c9fd14d2.pth - gpu_mem(M): '5420' - iter time(s): '3.27' - Config: configs/localization/bmn/bmn_400x100_2x8_9e_activitynet_feature.py In Collection: BMN Metadata: + Batch Size: 8 Epochs: 9 - Training BatchSize / GPU: 8 Training Data: ActivityNet v1.3 + Training Resources: 2 GPUs feature: mmaction_clip - gpus: 2 Name: bmn_400x100_9e_2x8_activitynet_feature (mmaction_clip) Results: - Dataset: ActivityNet v1.3 @@ -74,5 +68,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_2x8_9e_mmaction_clip/bmn_400x100_2x8_9e_mmaction_clip_20200809.json Training Log: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_2x8_9e_mmaction_clip/bmn_400x100_2x8_9e_mmaction_clip_20200809.log Weights: https://download.openmmlab.com/mmaction/localization/bmn/bmn_400x100_2x8_9e_mmaction_clip/bmn_400x100_2x8_9e_mmaction_clip_20200809-10d803ce.pth - gpu_mem(M): '5420' - iter time(s): '3.27' diff --git a/configs/localization/bsn/metafile.yml b/configs/localization/bsn/metafile.yml index bb64bda3b5..d3ccfbbb5d 100644 --- a/configs/localization/bsn/metafile.yml +++ b/configs/localization/bsn/metafile.yml @@ -1,7 +1,5 @@ Collections: -- Metadata: - Training Data: null - Name: BSN +- Name: BSN README: configs/localization/bsn/README.md Models: - Config: @@ -10,10 +8,10 @@ Models: - configs/localization/bsn/bsn_tem_400x100_1x16_20e_activitynet_feature.py In Collection: BSN Metadata: + Pretrained: None Training Data: ActivityNet v1.3 + Training Resources: 1 GPUs feature: cuhk_mean_100 - gpus: 1 - pretrain: None Name: bsn_400x100_1x16_20e_activitynet_feature (cuhk_mean_100) Results: - Dataset: ActivityNet v1.3 @@ -30,18 +28,16 @@ Models: Weights: - https://download.openmmlab.com/mmaction/localization/bsn/bsn_tem_400x100_1x16_20e_activitynet_feature/bsn_tem_400x100_1x16_20e_activitynet_feature_20200619-cd6accc3.pth - https://download.openmmlab.com/mmaction/localization/bsn/bsn_pem_400x100_1x16_20e_activitynet_feature/bsn_pem_400x100_1x16_20e_activitynet_feature_20210203-1c27763d.pth - gpu_mem(M): 41(TEM)+25(PEM) - iter time(s): 0.074(TEM)+0.036(PEM) - Config: - configs/localization/bsn/bsn_pem_400x100_1x16_20e_activitynet_feature.py - configs/localization/bsn/bsn_pgm_400x100_activitynet_feature.py - configs/localization/bsn/bsn_tem_400x100_1x16_20e_activitynet_feature.py In Collection: BSN Metadata: + Pretrained: None Training Data: ActivityNet v1.3 + Training Resources: 1 GPUs feature: mmaction_video - gpus: 1 - pretrain: None Name: bsn_400x100_1x16_20e_activitynet_feature (mmaction_video) Results: - Dataset: ActivityNet v1.3 @@ -58,18 +54,16 @@ Models: Weights: - https://download.openmmlab.com/mmaction/localization/bsn/bsn_tem_400x100_1x16_20e_mmaction_video/bsn_tem_400x100_1x16_20e_mmaction_video_20200809-ad6ec626.pth - https://download.openmmlab.com/mmaction/localization/bsn/bsn_pem_400x100_1x16_20e_mmaction_video/bsn_pem_400x100_1x16_20e_mmaction_video_20200809-aa861b26.pth - gpu_mem(M): 41(TEM)+25(PEM) - iter time(s): 0.074(TEM)+0.036(PEM) - Config: - configs/localization/bsn/bsn_pem_400x100_1x16_20e_activitynet_feature.py - configs/localization/bsn/bsn_pgm_400x100_activitynet_feature.py - configs/localization/bsn/bsn_tem_400x100_1x16_20e_activitynet_feature.py In Collection: BSN Metadata: + Pretrained: None Training Data: ActivityNet v1.3 + Training Resources: 1 GPUs feature: mmaction_clip - gpus: 1 - pretrain: None Name: bsn_400x100_1x16_20e_activitynet_feature (mmaction_clip) Results: - Dataset: ActivityNet v1.3 @@ -86,5 +80,3 @@ Models: Weights: - https://download.openmmlab.com/mmaction/localization/bsn/bsn_tem_400x100_1x16_20e_mmaction_clip/bsn_tem_400x100_1x16_20e_mmaction_clip_20200809-0a563554.pth - https://download.openmmlab.com/mmaction/localization/bsn/bsn_pem_400x100_1x16_20e_mmaction_clip/bsn_pem_400x100_1x16_20e_mmaction_clip_20200809-e32f61e6.pth - gpu_mem(M): 41(TEM)+25(PEM) - iter time(s): 0.074(TEM)+0.036(PEM) diff --git a/configs/localization/ssn/metafile.yml b/configs/localization/ssn/metafile.yml index 9cf416053a..29da43cabe 100644 --- a/configs/localization/ssn/metafile.yml +++ b/configs/localization/ssn/metafile.yml @@ -1,16 +1,14 @@ Collections: -- Metadata: - Training Data: null - Name: SSN +- Name: SSN README: configs/localization/ssn/README.md Models: - Config: configs/localization/ssn/ssn_r50_450e_thumos14_rgb_train.py In Collection: SSN Metadata: Architecture: ResNet50 + Pretrained: ImageNet Training Data: THUMOS 14 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Name: ssn_r50_450e_thumos14_rgb Results: - Dataset: THUMOS 14 @@ -22,7 +20,6 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/localization/ssn/ssn_r50_450e_thumos14_rgb/20201005_144656.log.json Training Log: https://download.openmmlab.com/mmaction/localization/ssn/ssn_r50_450e_thumos14_rgb/20201005_144656.log Weights: https://download.openmmlab.com/mmaction/localization/ssn/ssn_r50_450e_thumos14_rgb/ssn_r50_450e_thumos14_rgb_20201012-1920ab16.pth - gpu_mem(M): '6352' reference mAP@0.3: '[27.61](https://github.com/open-mmlab/mmaction/tree/c7e3b7c11fb94131be9b48a8e3d510589addc3ce#Get%20started)' reference mAP@0.4: '[21.28](https://github.com/open-mmlab/mmaction/tree/c7e3b7c11fb94131be9b48a8e3d510589addc3ce#Get%20started)' reference mAP@0.5: '[14.57](https://github.com/open-mmlab/mmaction/tree/c7e3b7c11fb94131be9b48a8e3d510589addc3ce#Get%20started)' diff --git a/configs/recognition/c3d/metafile.yml b/configs/recognition/c3d/metafile.yml index f0af255121..8dabc63831 100644 --- a/configs/recognition/c3d/metafile.yml +++ b/configs/recognition/c3d/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: C3D +- Name: C3D README: configs/recognition/c3d/README.md Models: - Config: configs/recognition/c3d/c3d_sports1m_16x1x1_45e_ucf101_rgb.py In Collection: C3D Metadata: Architecture: c3d + Batch Size: 30 Epochs: 45 Parameters: 78409573 - Training BatchSize / GPU: 30 + Pretrained: sports1m + Resolution: 128x171 Training Data: UCF101 - gpus: 8 - pretrain: sports1m - resolution: 128x171 + Training Resources: 8 GPUs Modality: RGB Name: c3d_sports1m_16x1x1_45e_ucf101_rgb.py Results: @@ -26,5 +24,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/c3d/c3d_sports1m_16x1x1_45e_ucf101_rgb/20201021_140429.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/c3d/c3d_sports1m_16x1x1_45e_ucf101_rgb/20201021_140429.log Weights: https://download.openmmlab.com/mmaction/recognition/c3d/c3d_sports1m_16x1x1_45e_ucf101_rgb/c3d_sports1m_16x1x1_45e_ucf101_rgb_20201021-26655025.pth - gpu_mem(M): '6053' - inference_time(video/s): x diff --git a/configs/recognition/csn/metafile.yml b/configs/recognition/csn/metafile.yml index 4c1cea7bf9..9dd6136cc2 100644 --- a/configs/recognition/csn/metafile.yml +++ b/configs/recognition/csn/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: CSN +- Name: CSN README: configs/recognition/csn/README.md Models: - Config: configs/recognition/csn/ircsn_ig65m_pretrained_r152_32x2x1_58e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet152 + Batch Size: 3 Epochs: 58 Parameters: 29703568 - Training BatchSize / GPU: 3 + Pretrained: IG65M + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: IG65M - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: ircsn_ig65m_pretrained_r152_32x2x1_58e_kinetics400_rgb.py Results: @@ -26,19 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/csn/ircsn_ig65m_pretrained_r152_32x2x1_58e_kinetics400_rgb/20200728_031952.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/csn/ircsn_ig65m_pretrained_r152_32x2x1_58e_kinetics400_rgb/20200728_031952.log Weights: https://download.openmmlab.com/mmaction/recognition/csn/ircsn_ig65m_pretrained_r152_32x2x1_58e_kinetics400_rgb/ircsn_ig65m_pretrained_r152_32x2x1_58e_kinetics400_rgb_20200803-fc66ce8d.pth - gpu_mem(M): '8517' - inference_time(video/s): x - Config: configs/recognition/csn/ircsn_ig65m_pretrained_bnfrozen_r152_32x2x1_58e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet152 + Batch Size: 3 Epochs: 58 Parameters: 29703568 - Training BatchSize / GPU: 3 + Pretrained: IG65M + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: IG65M - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: ircsn_ig65m_pretrained_bnfrozen_r152_32x2x1_58e_kinetics400_rgb.py Results: @@ -50,5 +46,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/csn/ircsn_ig65m_pretrained_bnfrozen_r152_32x2x1_58e_kinetics400_rgb/20200809_053132.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/csn/ircsn_ig65m_pretrained_bnfrozen_r152_32x2x1_58e_kinetics400_rgb/20200809_053132.log Weights: https://download.openmmlab.com/mmaction/recognition/csn/ircsn_ig65m_pretrained_bnfrozen_r152_32x2x1_58e_kinetics400_rgb/ircsn_ig65m_pretrained_bnfrozen_r152_32x2x1_58e_kinetics400_rgb_20200812-9037a758.pth - gpu_mem(M): '8516' - inference_time(video/s): x diff --git a/configs/recognition/i3d/metafile.yml b/configs/recognition/i3d/metafile.yml index 8f2f63c730..5a32517893 100644 --- a/configs/recognition/i3d/metafile.yml +++ b/configs/recognition/i3d/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: I3D +- Name: I3D README: configs/recognition/i3d/README.md Models: - Config: configs/recognition/i3d/i3d_r50_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 8 GPUs Modality: RGB Name: i3d_r50_32x2x1_100e_kinetics400_rgb Results: @@ -26,19 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_32x2x1_100e_kinetics400_rgb/20200614_060456.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_32x2x1_100e_kinetics400_rgb/20200614_060456.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_32x2x1_100e_kinetics400_rgb/i3d_r50_32x2x1_100e_kinetics400_rgb_20200614-c25ef9a4.pth - gpu_mem(M): '5170' - inference_time(video/s): 1.7 (320x3 frames) - Config: configs/recognition/i3d/i3d_r50_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: i3d_r50_32x2x1_100e_kinetics400_rgb Results: @@ -50,19 +46,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_256p_32x2x1_100e_kinetics400_rgb/20200725_031555.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_256p_32x2x1_100e_kinetics400_rgb/20200725_031555.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_256p_32x2x1_100e_kinetics400_rgb/i3d_r50_256p_32x2x1_100e_kinetics400_rgb_20200801-7d9f44de.pth - gpu_mem(M): '5170' - inference_time(video/s): x - Config: configs/recognition/i3d/i3d_r50_video_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256p Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256p + Training Resources: 8 GPUs Modality: RGB Name: i3d_r50_video_32x2x1_100e_kinetics400_rgb Results: @@ -74,19 +68,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_video_32x2x1_100e_kinetics400_rgb/20200706_143014.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_video_32x2x1_100e_kinetics400_rgb/20200706_143014.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_video_32x2x1_100e_kinetics400_rgb/i3d_r50_video_32x2x1_100e_kinetics400_rgb_20200826-e31c6f52.pth - gpu_mem(M): '5170' - inference_time(video/s): x - Config: configs/recognition/i3d/i3d_r50_dense_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 16 GPUs Modality: RGB Name: i3d_r50_dense_32x2x1_100e_kinetics400_rgb Results: @@ -98,19 +90,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_dense_32x2x1_100e_kinetics400_rgb/20200616_230011.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_dense_32x2x1_100e_kinetics400_rgb/20200616_230011.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_dense_32x2x1_100e_kinetics400_rgb/i3d_r50_dense_32x2x1_100e_kinetics400_rgb_20200616-2bbb4361.pth - gpu_mem(M): '5170' - inference_time(video/s): 1.7 (320x3 frames) - Config: configs/recognition/i3d/i3d_r50_dense_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: i3d_r50_dense_32x2x1_100e_kinetics400_rgb Results: @@ -122,19 +112,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_dense_256p_32x2x1_100e_kinetics400_rgb/20200725_031604.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_dense_256p_32x2x1_100e_kinetics400_rgb/20200725_031604.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_dense_256p_32x2x1_100e_kinetics400_rgb/i3d_r50_dense_256p_32x2x1_100e_kinetics400_rgb_20200725-24eb54cc.pth - gpu_mem(M): '5170' - inference_time(video/s): x - Config: configs/recognition/i3d/i3d_r50_lazy_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 8 GPUs Modality: RGB Name: i3d_r50_lazy_32x2x1_100e_kinetics400_rgb Results: @@ -146,19 +134,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_fast_32x2x1_100e_kinetics400_rgb/20200612_233836.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_fast_32x2x1_100e_kinetics400_rgb/20200612_233836.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_fast_32x2x1_100e_kinetics400_rgb/i3d_r50_fast_32x2x1_100e_kinetics400_rgb_20200612-000e4d2a.pth - gpu_mem(M): '5170' - inference_time(video/s): 1.8 (320x3 frames) - Config: configs/recognition/i3d/i3d_r50_lazy_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 28043472 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: i3d_r50_lazy_32x2x1_100e_kinetics400_rgb Results: @@ -170,19 +156,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_fast_256p_32x2x1_100e_kinetics400_rgb/20200725_031457.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_fast_256p_32x2x1_100e_kinetics400_rgb/20200725_031457.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_r50_fast_256p_32x2x1_100e_kinetics400_rgb/i3d_r50_fast_256p_32x2x1_100e_kinetics400_rgb_20200817-4e90d1d5.pth - gpu_mem(M): '5170' - inference_time(video/s): x - Config: configs/recognition/i3d/i3d_nl_embedded_gaussian_r50_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 35397840 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256p Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 256p + Training Resources: 32 GPUs Modality: RGB Name: i3d_nl_embedded_gaussian_r50_32x2x1_100e_kinetics400_rgb Results: @@ -194,19 +178,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_embedded_gaussian_r50_32x2x1_100e_kinetics400_rgb/20200813_034054.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_embedded_gaussian_r50_32x2x1_100e_kinetics400_rgb/20200813_034054.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_embedded_gaussian_r50_32x2x1_100e_kinetics400_rgb/i3d_nl_embedded_gaussian_r50_32x2x1_100e_kinetics400_rgb_20200813-6e6aef1b.pth - gpu_mem(M): '6438' - inference_time(video/s): x - Config: configs/recognition/i3d/i3d_nl_gaussian_r50_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 31723728 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256p Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 256p + Training Resources: 32 GPUs Modality: RGB Name: i3d_nl_gaussian_r50_32x2x1_100e_kinetics400_rgb Results: @@ -218,19 +200,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_gaussian_r50_32x2x1_100e_kinetics400_rgb/20200813_034909.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_gaussian_r50_32x2x1_100e_kinetics400_rgb/20200813_034909.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_gaussian_r50_32x2x1_100e_kinetics400_rgb/i3d_nl_gaussian_r50_32x2x1_100e_kinetics400_rgb_20200815-17f84aa2.pth - gpu_mem(M): '4944' - inference_time(video/s): x - Config: configs/recognition/i3d/i3d_nl_dot_product_r50_32x2x1_100e_kinetics400_rgb.py In Collection: CSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 35397840 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256p Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 256p + Training Resources: 32 GPUs Modality: RGB Name: i3d_nl_dot_product_r50_32x2x1_100e_kinetics400_rgb Results: @@ -242,5 +222,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_dot_product_r50_32x2x1_100e_kinetics400_rgb/20200814_044208.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_dot_product_r50_32x2x1_100e_kinetics400_rgb/20200814_044208.log Weights: https://download.openmmlab.com/mmaction/recognition/i3d/i3d_nl_dot_product_r50_32x2x1_100e_kinetics400_rgb/i3d_nl_dot_product_r50_32x2x1_100e_kinetics400_rgb_20200814-7c30d5bb.pth - gpu_mem(M): '4832' - inference_time(video/s): x diff --git a/configs/recognition/omnisource/metafile.yml b/configs/recognition/omnisource/metafile.yml index e8b6cf9758..16ca8a1896 100644 --- a/configs/recognition/omnisource/metafile.yml +++ b/configs/recognition/omnisource/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: OmniSource +- Name: OmniSource README: configs/recognition/omnisource/README.md Models: - Config: configs/recognition/omnisource/tsn_r50_1x1x8_100e_minikinetics/tsn_r50_1x1x8_100e_minikinetics_rgb.py In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Input: 3seg Modality: RGB Parameters: 23917832 Pretrained: ImageNet Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: tsn_r50_1x1x8_100e_minikinetics_rgb @@ -31,13 +29,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Input: 3seg Modality: RGB Parameters: 23917832 Pretrained: ImageNet Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: tsn_r50_1x1x8_100e_minikinetics_googleimage_rgb @@ -54,13 +52,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Input: 3seg Modality: RGB Parameters: 23917832 Pretrained: ImageNet Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: tsn_r50_1x1x8_100e_minikinetics_webimage_rgb @@ -77,13 +75,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Input: 3seg Modality: RGB Parameters: 23917832 Pretrained: ImageNet Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: tsn_r50_1x1x8_100e_minikinetics_insvideo_rgb @@ -100,13 +98,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Input: 3seg Modality: RGB Parameters: 23917832 Pretrained: ImageNet Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: tsn_r50_1x1x8_100e_minikinetics_kineticsraw_rgb @@ -123,13 +121,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Input: 3seg Modality: RGB Parameters: 23917832 Pretrained: ImageNet Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: tsn_r50_1x1x8_100e_minikinetics_omnisource_rgb @@ -146,13 +144,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Input: 8x8 Modality: RGB Parameters: 32044296 Pretrained: None Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: slowonly_r50_8x8x1_256e_minikinetics_rgb @@ -169,13 +167,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Input: 8x8 Modality: RGB Parameters: 32044296 Pretrained: None Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: slowonly_r50_8x8x1_256e_minikinetics_googleimage_rgb @@ -192,13 +190,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Input: 8x8 Modality: RGB Parameters: 32044296 Pretrained: None Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: slowonly_r50_8x8x1_256e_minikinetics_webimage_rgb @@ -215,13 +213,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Input: 8x8 Modality: RGB Parameters: 32044296 Pretrained: None Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: slowonly_r50_8x8x1_256e_minikinetics_insvideo_rgb @@ -238,13 +236,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Input: 8x8 Modality: RGB Parameters: 32044296 Pretrained: None Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: slowonly_r50_8x8x1_256e_minikinetics_kineticsraw_rgb @@ -261,13 +259,13 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Input: 8x8 Modality: RGB Parameters: 32044296 Pretrained: None Resolution: short-side 320 - Training BatchSize / GPU: 12 Training Data: MiniKinetics Modality: RGB Name: slowonly_r50_8x8x1_256e_minikinetics_omnisource_rgb @@ -284,12 +282,12 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - pretrain: ImageNet - resolution: 340x256 Modality: RGB Name: tsn_omnisource_r50_1x1x3_100e_kinetics_rgb Results: @@ -299,18 +297,16 @@ Models: top5 acc: 91.0 Task: Action Recognition Weights: https://download.openmmlab.com/mmaction/recognition/tsn/omni/tsn_imagenet_pretrained_r50_omni_1x1x3_kinetics400_rgb_20200926-54192355.pth - gpu_mem(M): '8344' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb.py In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: IG-1B + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: IG-1B - resolution: short-side 320 Modality: RGB Name: tsn_IG1B_pretrained_omnisource_r50_1x1x3_100e_kinetics_rgb Results: @@ -320,18 +316,16 @@ Models: top5 acc: 91.9 Task: Action Recognition Weights: https://download.openmmlab.com/mmaction/recognition/tsn/omni/tsn_1G1B_pretrained_r50_omni_1x1x3_kinetics400_rgb_20200926-2863fed0.pth - gpu_mem(M): '8344' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_rgb.py In Collection: OmniSource Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: None - resolution: short-side 320 Modality: RGB Name: slowonly_r50_omnisource_4x16x1_256e_kinetics400_rgb Results: @@ -345,12 +339,12 @@ Models: In Collection: OmniSource Metadata: Architecture: ResNet101 + Batch Size: 8 Epochs: 196 Parameters: 60359120 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: None - resolution: short-side 320 Modality: RGB Name: slowonly_r101_omnisource_8x8x1_196e_kinetics400_rgb Results: diff --git a/configs/recognition/r2plus1d/metafile.yml b/configs/recognition/r2plus1d/metafile.yml index d6f7fdcb81..70f8aaee85 100644 --- a/configs/recognition/r2plus1d/metafile.yml +++ b/configs/recognition/r2plus1d/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: R2Plus1D +- Name: R2Plus1D README: configs/recognition/r2plus1d/README.md Models: - Config: configs/recognition/r2plus1d/r2plus1d_r34_8x8x1_180e_kinetics400_rgb.py In Collection: R2Plus1D Metadata: Architecture: ResNet34 + Batch Size: 8 Epochs: 180 Parameters: 63759281 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: r2plus1d_r34_8x8x1_180e_kinetics400_rgb Results: @@ -26,19 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_256p_8x8x1_180e_kinetics400_rgb/20200728_021421.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_256p_8x8x1_180e_kinetics400_rgb/20200728_021421.log Weights: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_256p_8x8x1_180e_kinetics400_rgb/r2plus1d_r34_256p_8x8x1_180e_kinetics400_rgb_20200729-aa94765e.pth - gpu_mem(M): '5019' - inference_time(video/s): x - Config: configs/recognition/r2plus1d/r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb.py In Collection: R2Plus1D Metadata: Architecture: ResNet34 + Batch Size: 16 Epochs: 180 Parameters: 63759281 - Training BatchSize / GPU: 16 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: None - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb Results: @@ -50,19 +46,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb/20200724_201360.log Training Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb/20200724_201360.log.json Weights: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb/r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb_20200826-ab35a529.pth - gpu_mem(M): '5019' - inference_time(video/s): x - Config: configs/recognition/r2plus1d/r2plus1d_r34_8x8x1_180e_kinetics400_rgb.py In Collection: R2Plus1D Metadata: Architecture: ResNet34 + Batch Size: 8 Epochs: 180 Parameters: 63759281 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: None - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: r2plus1d_r34_8x8x1_180e_kinetics400_rgb Results: @@ -74,19 +68,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_8x8x1_180e_kinetics400_rgb/r2plus1d_r34_8x8_69.58_88.36.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_8x8x1_180e_kinetics400_rgb/r21d_8x8.log Weights: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_8x8x1_180e_kinetics400_rgb/r2plus1d_r34_8x8x1_180e_kinetics400_rgb_20200618-3fce5629.pth - gpu_mem(M): '5019' - inference_time(video/s): 1.6 (80x3 frames) - Config: configs/recognition/r2plus1d/r2plus1d_r34_32x2x1_180e_kinetics400_rgb.py In Collection: R2Plus1D Metadata: Architecture: ResNet34 + Batch Size: 6 Epochs: 180 Parameters: 63759281 - Training BatchSize / GPU: 6 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: None - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: r2plus1d_r34_32x2x1_180e_kinetics400_rgb Results: @@ -98,5 +90,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_32x2x1_180e_kinetics400_rgb/r2plus1d_r34_32x2_74.6_91.6.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_32x2x1_180e_kinetics400_rgb/r21d_32x2.log Weights: https://download.openmmlab.com/mmaction/recognition/r2plus1d/r2plus1d_r34_32x2x1_180e_kinetics400_rgb/r2plus1d_r34_32x2x1_180e_kinetics400_rgb_20200618-63462eb3.pth - gpu_mem(M): '12975' - inference_time(video/s): 0.5 (320x3 frames) diff --git a/configs/recognition/slowfast/metafile.yml b/configs/recognition/slowfast/metafile.yml index dc32dac835..5821af9ce2 100644 --- a/configs/recognition/slowfast/metafile.yml +++ b/configs/recognition/slowfast/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: SlowFast +- Name: SlowFast README: configs/recognition/slowfast/README.md Models: - Config: configs/recognition/slowfast/slowfast_r50_4x16x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 34479288 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowfast_r50_4x16x1_256e_kinetics400_rgb Results: @@ -26,19 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_256p_4x16x1_256e_kinetics400_rgb/20200731_151706.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_256p_4x16x1_256e_kinetics400_rgb/20200731_151706.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_256p_4x16x1_256e_kinetics400_rgb/slowfast_r50_256p_4x16x1_256e_kinetics400_rgb_20200728-145f1097.pth - gpu_mem(M): '6203' - inference_time(video/s): x - Config: configs/recognition/slowfast/slowfast_r50_video_4x16x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 34479288 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: None - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: slowfast_r50_video_4x16x1_256e_kinetics400_rgb Results: @@ -50,19 +46,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_video_4x16x1_256e_kinetics400_rgb/20200812_160237.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_video_4x16x1_256e_kinetics400_rgb/20200812_160237.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_video_4x16x1_256e_kinetics400_rgb/slowfast_r50_video_4x16x1_256e_kinetics400_rgb_20200826-f85b90c5.pth - gpu_mem(M): '6203' - inference_time(video/s): x - Config: configs/recognition/slowfast/slowfast_r50_4x16x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 34479288 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 24 - pretrain: None - resolution: short-side 320 + Training Resources: 24 GPUs Modality: RGB Name: slowfast_r50_4x16x1_256e_kinetics400_rgb Results: @@ -74,19 +68,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_4x16x1_256e_kinetics400_rgb/20200704_232901.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_4x16x1_256e_kinetics400_rgb/20200704_232901.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_4x16x1_256e_kinetics400_rgb/slowfast_r50_4x16x1_256e_kinetics400_rgb_20200704-bcde7ed7.pth - gpu_mem(M): '6203' - inference_time(video/s): 1.6 ((32+4)x10x3 frames) - Config: configs/recognition/slowfast/slowfast_r50_8x8x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 34565560 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowfast_r50_8x8x1_256e_kinetics400_rgb Results: @@ -98,19 +90,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_256p_8x8x1_256e_kinetics400_rgb/20200731_151537.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_256p_8x8x1_256e_kinetics400_rgb/20200731_151537.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_256p_8x8x1_256e_kinetics400_rgb/slowfast_r50_256p_8x8x1_256e_kinetics400_rgb_20200810-863812c2.pth - gpu_mem(M): '9062' - inference_time(video/s): x - Config: configs/recognition/slowfast/slowfast_r50_8x8x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 34565560 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 24 - pretrain: None - resolution: short-side 320 + Training Resources: 24 GPUs Modality: RGB Name: slowfast_r50_8x8x1_256e_kinetics400_rgb Results: @@ -122,19 +112,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_8x8x1_256e_kinetics400_rgb/20200716_192653.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_8x8x1_256e_kinetics400_rgb/20200716_192653.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r50_8x8x1_256e_kinetics400_rgb/slowfast_r50_8x8x1_256e_kinetics400_rgb_20200716-73547d2b.pth - gpu_mem(M): '9062' - inference_time(video/s): 1.3 ((32+8)x10x3 frames) - Config: configs/recognition/slowfast/slowfast_r101_r50_4x16x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet101 + ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 62384312 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: None - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: slowfast_r101_r50_4x16x1_256e_kinetics400_rgb Results: @@ -146,19 +134,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r101_4x16x1_256e_kinetics400_rgb/20210118_133528.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r101_4x16x1_256e_kinetics400_rgb/20210118_133528.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r101_4x16x1_256e_kinetics400_rgb/slowfast_r101_4x16x1_256e_kinetics400_rgb_20210218-d8b58813.pth - gpu_mem(M): '16628' - inference_time(video/s): '' - Config: configs/recognition/slowfast/slowfast_r101_8x8x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet101 + Batch Size: 8 Epochs: 256 Parameters: 62912312 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowfast_r101_8x8x1_256e_kinetics400_rgb Results: @@ -170,19 +156,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r101_8x8x1_256e_kinetics400_rgb/20210218_121513.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r101_8x8x1_256e_kinetics400_rgb/20210218_121513.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r101_8x8x1_256e_kinetics400_rgb/slowfast_r101_8x8x1_256e_kinetics400_rgb_20210218-0dd54025.pth - gpu_mem(M): '25994' - inference_time(video/s): '' - Config: configs/recognition/slowfast/slowfast_r152_r50_4x16x1_256e_kinetics400_rgb.py In Collection: SlowFast Metadata: Architecture: ResNet152 + ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 84843704 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: None - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: slowfast_r152_r50_4x16x1_256e_kinetics400_rgb Results: @@ -194,5 +178,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r152_4x16x1_256e_kinetics400_rgb/20210122_131321.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r152_4x16x1_256e_kinetics400_rgb/20210122_131321.log Weights: https://download.openmmlab.com/mmaction/recognition/slowfast/slowfast_r152_4x16x1_256e_kinetics400_rgb/slowfast_r152_4x16x1_256e_kinetics400_rgb_20210122-bdeb6b87.pth - gpu_mem(M): '10077' - inference_time(video/s): '' diff --git a/configs/recognition/slowonly/metafile.yml b/configs/recognition/slowonly/metafile.yml index d2ca9732b8..da79055056 100644 --- a/configs/recognition/slowonly/metafile.yml +++ b/configs/recognition/slowonly/metafile.yml @@ -1,19 +1,17 @@ Collections: -- Metadata: - Training Data: null - Name: SlowOnly +- Name: SlowOnly README: configs/recognition/slowonly/README.md Models: - Config: configs/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: None - resolution: short-side 320 Modality: RGB Name: slowonly_r50_omnisource_4x16x1_256e_kinetics400_rgb Results: @@ -27,12 +25,12 @@ Models: In Collection: SlowOnly Metadata: Architecture: ResNet101 + Batch Size: 8 Epochs: 196 Parameters: 60359120 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: None - resolution: short-side 320 Modality: RGB Name: slowonly_r101_8x8x1_196e_kinetics400_rgb Results: @@ -46,12 +44,12 @@ Models: In Collection: SlowOnly Metadata: Architecture: ResNet101 + Batch Size: 8 Epochs: 196 Parameters: 60359120 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: None - resolution: short-side 320 Modality: RGB Name: slowonly_r101_omnisource_8x8x1_196e_kinetics400_rgb Results: @@ -65,13 +63,13 @@ Models: In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowonly_r50_4x16x1_256e_kinetics400_rgb Results: @@ -83,19 +81,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_256p_4x16x1_256e_kinetics400_rgb/20200817_001411.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_256p_4x16x1_256e_kinetics400_rgb/20200817_001411.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_256p_4x16x1_256e_kinetics400_rgb/slowonly_r50_256p_4x16x1_256e_kinetics400_rgb_20200820-bea7701f.pth - gpu_mem(M): '3168' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_video_4x16x1_256e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 24 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 24 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: None - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_r50_video_4x16x1_256e_kinetics400_rgb Results: @@ -107,19 +103,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_320p_4x16x1_256e_kinetics400_rgb/slowonly_r50_video_320p_4x16x1_256e_kinetics400_rgb_20201014.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_320p_4x16x1_256e_kinetics400_rgb/slowonly_r50_video_320p_4x16x1_256e_kinetics400_rgb_20201014.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_video_320p_4x16x1_256e_kinetics400_rgb/slowonly_r50_video_320p_4x16x1_256e_kinetics400_rgb_20201014-c9cdc656.pth - gpu_mem(M): '8472' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowonly_r50_8x8x1_256e_kinetics400_rgb Results: @@ -131,19 +125,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_256p_8x8x1_256e_kinetics400_rgb/20200817_003320.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_256p_8x8x1_256e_kinetics400_rgb/20200817_003320.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_256p_8x8x1_256e_kinetics400_rgb/slowonly_r50_256p_8x8x1_256e_kinetics400_rgb_20200820-75851a7d.pth - gpu_mem(M): '5820' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: None - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_r50_4x16x1_256e_kinetics400_rgb Results: @@ -155,19 +147,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_rgb/slowonly_r50_4x16_73.02_90.77.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_rgb/so_4x16.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_rgb/slowonly_r50_4x16x1_256e_kinetics400_rgb_20200704-a69556c6.pth - gpu_mem(M): '3168' - inference_time(video/s): 4.0 (40x3 frames) - Config: configs/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 256 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: None + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 24 - pretrain: None - resolution: short-side 320 + Training Resources: 24 GPUs Modality: RGB Name: slowonly_r50_8x8x1_256e_kinetics400_rgb Results: @@ -179,19 +169,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_rgb/slowonly_r50_8x8_74.93_91.92.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_rgb/so_8x8.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_rgb/slowonly_r50_8x8x1_256e_kinetics400_rgb_20200703-a79c555a.pth - gpu_mem(M): '5820' - inference_time(video/s): 2.3 (80x3 frames) - Config: configs/recognition/slowonly/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb Results: @@ -203,19 +191,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb_20200912.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb_20200912.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb/slowonly_imagenet_pretrained_r50_4x16x1_150e_kinetics400_rgb_20200912-1e8fc736.pth - gpu_mem(M): '3168' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 32454096 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb Results: @@ -227,19 +213,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb_20200912.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb_20200912.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb/slowonly_imagenet_pretrained_r50_8x8x1_150e_kinetics400_rgb_20200912-3f9ce182.pth - gpu_mem(M): '5820' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_4x16x1_150e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 39808464 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_nl_embedded_gaussian_r50_4x16x1_150e_kinetics400_rgb Results: @@ -251,19 +235,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_4x16x1_150e_kinetics400_rgb/20210305_152630.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_4x16x1_150e_kinetics400_rgb/20210305_152630.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_4x16x1_150e_kinetics400_rgb/slowonly_nl_embedded_gaussian_r50_4x16x1_150e_kinetics400_rgb_20210308-0d6e5a69.pth - gpu_mem(M): '4435' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_8x8x1_150e_kinetics400_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 39808464 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: slowonly_nl_embedded_gaussian_r50_8x8x1_150e_kinetics400_rgb Results: @@ -275,19 +257,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_8x8x1_150e_kinetics400_rgb/20210308_212250.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_8x8x1_150e_kinetics400_rgb/20210308_212250.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_nl_embedded_gaussian_r50_8x8x1_150e_kinetics400_rgb/slowonly_nl_embedded_gaussian_r50_8x8x1_150e_kinetics400_rgb_20210308-e8dd9e82.pth - gpu_mem(M): '8895' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_flow.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 24 Epochs: 256 Parameters: 32450960 - Training BatchSize / GPU: 24 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: Flow Name: slowonly_r50_4x16x1_256e_kinetics400_flow Results: @@ -299,19 +279,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_flow/slowonly_r50_4x16x1_256e_kinetics400_flow_61.8_83.6.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_flow/slowonly_r50_4x16x1_256e_kinetics400_flow_61.8_83.6.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_4x16x1_256e_kinetics400_flow/slowonly_r50_4x16x1_256e_kinetics400_flow_20200704-decb8568.pth - gpu_mem(M): '8450' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_flow.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 196 Parameters: 32450960 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: Flow Name: slowonly_r50_8x8x1_196e_kinetics400_flow Results: @@ -323,19 +301,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_flow/slowonly_r50_8x8x1_196e_kinetics400_flow_65.8_86.3.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_flow/slowonly_r50_8x8x1_196e_kinetics400_flow_65.8_86.3.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_r50_8x8x1_256e_kinetics400_flow/slowonly_r50_8x8x1_256e_kinetics400_flow_20200704-6b384243.pth - gpu_mem(M): '8455' - inference_time(video/s): x - Config: configs/recognition/slowonly/slowonly_r50_video_8x8x1_256e_kinetics600_rgb.py In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Parameters: 32863896 - Training BatchSize / GPU: 12 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-600 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowonly_r50_video_8x8x1_256e_kinetics600_rgb Results: @@ -351,13 +327,13 @@ Models: In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 256 Parameters: 33068796 - Training BatchSize / GPU: 12 + Pretrained: None + Resolution: short-side 256 Training Data: Kinetics-700 - gpus: 32 - pretrain: None - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: slowonly_r50_video_8x8x1_256e_kinetics700_rgb Results: @@ -373,13 +349,13 @@ Models: In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 24 Epochs: 120 Parameters: 32454096 - Training BatchSize / GPU: 24 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: GYM99 - gpus: 16 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 16 GPUs Modality: RGB Name: slowonly_imagenet_pretrained_r50_4x16x1_120e_gym99_rgb Results: @@ -395,13 +371,13 @@ Models: In Collection: SlowOnly Metadata: Architecture: ResNet50 + Batch Size: 24 Epochs: 120 Parameters: 32450960 - Training BatchSize / GPU: 24 + Pretrained: Kinetics + Resolution: short-side 256 Training Data: GYM99 - gpus: 16 - pretrain: Kinetics - resolution: short-side 256 + Training Resources: 16 GPUs Modality: Flow Name: slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow Results: @@ -413,3 +389,24 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow/slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow_20201111.json Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow/slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow_20201111.log Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow/slowonly_kinetics_pretrained_r50_4x16x1_120e_gym99_flow_20201111-66ecdb3c.pth +- Config: configs/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb.py + In Collection: SlowOnly + Metadata: + Architecture: ResNet50 + Batch Size: 8 + Epochs: 64 + Parameters: 31689819 + Pretrained: ImageNet + Resolution: height 100 + Training Data: Jester + Training Resources: 8 GPUs + Modality: RGB + Name: slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb + Results: + - Dataset: Jester + Metrics: + top1 acc: 97.2 + Task: Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb.log.json + Training Log: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb.log + Weights: https://download.openmmlab.com/mmaction/recognition/slowonly/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb/slowonly_imagenet_pretrained_r50_8x8x1_64e_jester_rgb-b56a5389.pth diff --git a/configs/recognition/tanet/metafile.yml b/configs/recognition/tanet/metafile.yml index 7f991f2ffc..6ce789d4ac 100644 --- a/configs/recognition/tanet/metafile.yml +++ b/configs/recognition/tanet/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: TANet +- Name: TANet README: configs/recognition/tanet/README.md Models: - Config: configs/recognition/tanet/tanet_r50_dense_1x1x8_100e_kinetics400_rgb.py In Collection: TANet Metadata: Architecture: TANet + Batch Size: 8 Epochs: 100 Parameters: 25590320 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 8 GPUs Modality: RGB Name: tanet_r50_dense_1x1x8_100e_kinetics400_rgb Results: @@ -26,5 +24,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tanet/tanet_r50_dense_1x1x8_100e_kinetics400_rgb/tanet_r50_dense_1x1x8_100e_kinetics400_rgb_20210219.json Training Log: https://download.openmmlab.com/mmaction/recognition/tanet/tanet_r50_dense_1x1x8_100e_kinetics400_rgb/tanet_r50_dense_1x1x8_100e_kinetics400_rgb_20210219.log Weights: https://download.openmmlab.com/mmaction/recognition/tanet/tanet_r50_dense_1x1x8_100e_kinetics400_rgb/tanet_r50_dense_1x1x8_100e_kinetics400_rgb_20210219-032c8e94.pth - gpu_mem(M): '7124' - inference_time(video/s): x diff --git a/configs/recognition/tin/metafile.yml b/configs/recognition/tin/metafile.yml index e765a40046..f67be352d9 100644 --- a/configs/recognition/tin/metafile.yml +++ b/configs/recognition/tin/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: TIN +- Name: TIN README: configs/recognition/tin/README.md Models: - Config: configs/recognition/tin/tin_r50_1x1x8_40e_sthv1_rgb.py In Collection: TIN Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 40 Parameters: 23895566 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 32 - pretrain: ImageNet - resolution: height 100 + Training Resources: 32 GPUs Modality: RGB Name: tin_r50_1x1x8_40e_sthv1_rgb Results: @@ -26,18 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tin/tin_r50_1x1x8_40e_sthv1_rgb/20200729_034132.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tin/tin_r50_1x1x8_40e_sthv1_rgb/20200729_034132.log Weights: https://download.openmmlab.com/mmaction/recognition/tin/tin_r50_1x1x8_40e_sthv1_rgb/tin_r50_1x1x8_40e_sthv1_rgb_20200729-4a33db86.pth - gpu_mem(M): '6181' - Config: configs/recognition/tin/tin_r50_1x1x8_40e_sthv2_rgb.py In Collection: TIN Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 40 Parameters: 23895566 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 240 Training Data: SthV2 - gpus: 32 - pretrain: ImageNet - resolution: height 240 + Training Resources: 32 GPUs Modality: RGB Name: tin_r50_1x1x8_40e_sthv2_rgb Results: @@ -49,18 +46,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tin/tin_r50_1x1x8_40e_sthv2_rgb/20200912_225451.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tin/tin_r50_1x1x8_40e_sthv2_rgb/20200912_225451.log Weights: https://download.openmmlab.com/mmaction/recognition/tin/tin_r50_1x1x8_40e_sthv2_rgb/tin_r50_1x1x8_40e_sthv2_rgb_20200912-b27a7337.pth - gpu_mem(M): '6185' - Config: configs/recognition/tin/tin_tsm_finetune_r50_1x1x8_50e_kinetics400_rgb.py In Collection: TIN Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 24358640 - Training BatchSize / GPU: 6 + Pretrained: TSM-Kinetics400 + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: TSM-Kinetics400 - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: tin_tsm_finetune_r50_1x1x8_50e_kinetics400_rgb Results: @@ -72,4 +68,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tin/tin_tsm_finetune_r50_1x1x8_50e_kinetics400_rgb/20200809_142447.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tin/tin_tsm_finetune_r50_1x1x8_50e_kinetics400_rgb/20200809_142447.log Weights: https://download.openmmlab.com/mmaction/recognition/tin/tin_tsm_finetune_r50_1x1x8_50e_kinetics400_rgb/tin_tsm_finetune_r50_1x1x8_50e_kinetics400_rgb_20200810-4a146a70.pth - gpu_mem(M): '6187' diff --git a/configs/recognition/tpn/metafile.yml b/configs/recognition/tpn/metafile.yml index ca0655ff49..b091904d46 100644 --- a/configs/recognition/tpn/metafile.yml +++ b/configs/recognition/tpn/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: TPN +- Name: TPN README: configs/recognition/tpn/README.md Models: - Config: configs/recognition/tpn/tpn_slowonly_r50_8x8x1_150e_kinetics_rgb.py In Collection: TPN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 91498336 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: tpn_slowonly_r50_8x8x1_150e_kinetics_rgb Results: @@ -26,19 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_slowonly_r50_8x8x1_150e_kinetics_rgb/20200910_134330.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_slowonly_r50_8x8x1_150e_kinetics_rgb/20200910_134330.log Weights: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_slowonly_r50_8x8x1_150e_kinetics_rgb/tpn_slowonly_r50_8x8x1_150e_kinetics_rgb_20200910-b796d7a0.pth - gpu_mem(M): '6916' - inference_time(video/s): x - Config: configs/recognition/tpn/tpn_imagenet_pretrained_slowonly_r50_8x8x1_150e_kinetics_rgb.py In Collection: TPN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 91498336 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: tpn_imagenet_pretrained_slowonly_r50_8x8x1_150e_kinetics_rgb Results: @@ -50,19 +46,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_imagenet_pretrained_slowonly_r50_8x8x1_150e_kinetics_rgb/20200923_151919.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_imagenet_pretrained_slowonly_r50_8x8x1_150e_kinetics_rgb/20200923_151919.log Weights: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_imagenet_pretrained_slowonly_r50_8x8x1_150e_kinetics_rgb/tpn_imagenet_pretrained_slowonly_r50_8x8x1_150e_kinetics_rgb_20200923-52629684.pth - gpu_mem(M): '6916' - inference_time(video/s): x - Config: configs/recognition/tpn/tpn_tsm_r50_1x1x8_150e_sthv1_rgb.py In Collection: TPN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 82445724 - Training BatchSize / GPU: 8 + Pretrained: TSM + Resolution: height 100 Training Data: SthV1 - gpus: 48 - pretrain: TSM - resolution: height 100 + Training Resources: 48 GPUs Modality: RGB Name: tpn_tsm_r50_1x1x8_150e_sthv1_rgb Results: @@ -74,4 +68,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_tsm_r50_1x1x8_150e_sthv1_rgb/20210311_162636.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_tsm_r50_1x1x8_150e_sthv1_rgb/20210311_162636.log Weights: https://download.openmmlab.com/mmaction/recognition/tpn/tpn_tsm_r50_1x1x8_150e_sthv1_rgb/tpn_tsm_r50_1x1x8_150e_sthv1_rgb_20210311-28de4cd5.pth - gpu_mem(M): '8828' diff --git a/configs/recognition/trn/metafile.yml b/configs/recognition/trn/metafile.yml index c4e5586939..10890f14de 100644 --- a/configs/recognition/trn/metafile.yml +++ b/configs/recognition/trn/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: TRN +- Name: TRN README: configs/recognition/trn/README.md Models: - Config: configs/recognition/trn/trn_r50_1x1x8_50e_sthv1_rgb.py In Collection: TRN Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 50 Parameters: 26641154 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: trn_r50_1x1x8_50e_sthv1_rgb Results: @@ -28,18 +26,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/trn/trn_r50_1x1x8_50e_sthv1_rgb/20210326_103948.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/trn/trn_r50_1x1x8_50e_sthv1_rgb/20210326_103948.log Weights: https://download.openmmlab.com/mmaction/recognition/trn/trn_r50_1x1x8_50e_sthv1_rgb/trn_r50_1x1x8_50e_sthv1_rgb_20210401-163704a8.pth - gpu_mem(M): '11010' - Config: configs/recognition/trn/trn_r50_1x1x8_50e_sthv2_rgb.py In Collection: TRN Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 50 Parameters: 26641154 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: trn_r50_1x1x8_50e_sthv2_rgb Results: @@ -53,4 +50,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/trn/trn_r50_1x1x8_50e_sthv2_rgb/20210326_103951.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/trn/trn_r50_1x1x8_50e_sthv2_rgb/20210326_103951.log Weights: https://download.openmmlab.com/mmaction/recognition/trn/trn_r50_1x1x8_50e_sthv2_rgb/trn_r50_1x1x8_50e_sthv2_rgb_20210401-773eca7b.pth - gpu_mem(M): '11010' diff --git a/configs/recognition/tsm/metafile.yml b/configs/recognition/tsm/metafile.yml index 6c099238aa..11bd167de2 100644 --- a/configs/recognition/tsm/metafile.yml +++ b/configs/recognition/tsm/metafile.yml @@ -1,20 +1,18 @@ Collections: -- Metadata: - Training Data: null - Name: TSM +- Name: TSM README: configs/recognition/tsm/README.md Models: - Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 24327632 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x8_50e_kinetics400_rgb Results: @@ -26,19 +24,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb/20200607_211800.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb/20200607_211800.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb/tsm_r50_1x1x8_50e_kinetics400_rgb_20200607-af7fb746.pth - gpu_mem(M): '7079' - inference_time(video/s): 74.0 (8x1 frames) - Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 24327632 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x8_50e_kinetics400_rgb Results: @@ -50,19 +46,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x8_50e_kinetics400_rgb/20200725_031623.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x8_50e_kinetics400_rgb/20200725_031623.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x8_50e_kinetics400_rgb/tsm_r50_256p_1x1x8_50e_kinetics400_rgb_20200726-020785e2.pth - gpu_mem(M): '7079' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 24327632 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py Results: @@ -74,19 +68,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb_20210219.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb_20210219.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb_20210219-bf96e6cc.pth - gpu_mem(M): '7076' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_r50_video_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 24327632 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_video_1x1x8_50e_kinetics400_rgb Results: @@ -98,19 +90,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_100e_kinetics400_rgb/tsm_r50_video_2d_1x1x8_50e_kinetics400_rgb.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_100e_kinetics400_rgb/tsm_r50_video_2d_1x1x8_50e_kinetics400_rgb.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_100e_kinetics400_rgb/tsm_r50_video_1x1x8_100e_kinetics400_rgb_20200702-a77f4328.pth - gpu_mem(M): '7077' - inference_time(video/s): 74.0 (8x1 frames) - Config: configs/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 32 GPUs Modality: RGB Name: tsm_r50_dense_1x1x8_100e_kinetics400_rgb Results: @@ -122,19 +112,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/20200626_213415.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/20200626_213415.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/tsm_r50_dense_1x1x8_100e_kinetics400_rgb_20200626-91a54551.pth - gpu_mem(M): '7079' - inference_time(video/s): 11.5 (8x10 frames) - Config: configs/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_dense_1x1x8_100e_kinetics400_rgb Results: @@ -146,19 +134,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_256p_1x1x8_100e_kinetics400_rgb/20200725_032043.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_256p_1x1x8_100e_kinetics400_rgb/20200725_032043.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_256p_1x1x8_100e_kinetics400_rgb/tsm_r50_dense_256p_1x1x8_100e_kinetics400_rgb_20200727-e1e0c785.pth - gpu_mem(M): '7079' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 24327632 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x16_50e_kinetics400_rgb Results: @@ -170,19 +156,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb/20201011_205356.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb/20201011_205356.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb/tsm_r50_340x256_1x1x16_50e_kinetics400_rgb_20201011-2f27f229.pth - gpu_mem(M): '10404' - inference_time(video/s): 47.0 (16x1 frames) - Config: configs/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 24327632 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 32 GPUs Modality: RGB Name: tsm_r50_1x1x16_50e_kinetics400_rgb Results: @@ -194,19 +178,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x16_50e_kinetics400_rgb/20201010_224825.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x16_50e_kinetics400_rgb/20201010_224825.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x16_50e_kinetics400_rgb/tsm_r50_256p_1x1x16_50e_kinetics400_rgb_20201010-85645c2a.pth - gpu_mem(M): '10398' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 31682000 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb Results: @@ -218,19 +200,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200724_120023.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200724_120023.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb_20200724-f00f1336.pth - gpu_mem(M): '8931' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 28007888 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb Results: @@ -242,19 +222,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200815_210253.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200815_210253.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb_20200816-b93fd297.pth - gpu_mem(M): '10125' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 31682000 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: RGB Name: tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb Results: @@ -266,19 +244,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb/20200723_220442.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb/20200723_220442.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb_20200724-d8ad84d2.pth - gpu_mem(M): '8358' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb.py In Collection: TSM Metadata: Architecture: MobileNetV2 + Batch Size: 8 Epochs: 100 Parameters: 2736272 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 8 GPUs Modality: RGB Name: tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb Results: @@ -290,18 +266,16 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb/20210129_024936.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb/20210129_024936.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb/tsm_mobilenetv2_dense_320p_1x1x8_100e_kinetics400_rgb_20210202-61135809.pth - gpu_mem(M): '3385' - inference_time(video/s): x - Config: configs/recognition/tsm/tsm_r50_video_1x1x8_50e_diving48_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23606384 - Training BatchSize / GPU: 8 + Pretrained: ImageNet Training Data: Diving48 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_video_1x1x8_50e_diving48_rgb Results: @@ -313,17 +287,16 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_50e_diving48_rgb/20210426_012424.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_50e_diving48_rgb/20210426_012424.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_50e_diving48_rgb/tsm_r50_video_1x1x8_50e_diving48_rgb_20210426-aba5aa3d.pth - gpu_mem(M): '7070' - Config: configs/recognition/tsm/tsm_r50_video_1x1x16_50e_diving48_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 4 Epochs: 50 Parameters: 23606384 - Training BatchSize / GPU: 4 + Pretrained: ImageNet Training Data: Diving48 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_video_1x1x16_50e_diving48_rgb Results: @@ -335,18 +308,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x16_50e_diving48_rgb/20210426_012823.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x16_50e_diving48_rgb/20210426_012823.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x16_50e_diving48_rgb/tsm_r50_video_1x1x16_50e_diving48_rgb_20210426-aa9631c0.pth - gpu_mem(M): '7070' - Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x8_50e_sthv1_rgb Results: @@ -360,20 +332,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb/20210203_150227.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb/20210203_150227.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb/tsm_r50_1x1x8_50e_sthv1_rgb_20210203-01dce462.pth - gpu_mem(M): '7077' reference top1 acc (efficient/accurate): '[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_flip_1x1x8_50e_sthv1_rgb Results: @@ -387,20 +358,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb/20210203_145829.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb/20210203_145829.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb/tsm_r50_flip_1x1x8_50e_sthv1_rgb_20210203-12596f16.pth - gpu_mem(M): '7077' reference top1 acc (efficient/accurate): '[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_randaugment_1x1x8_50e_sthv1_rgb Results: @@ -414,20 +384,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb_20210324-481268d9.pth - gpu_mem(M): '7077' reference top1 acc (efficient/accurate): '[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb Results: @@ -441,20 +410,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb_20210324-76937692.pth - gpu_mem(M): '7077' reference top1 acc (efficient/accurate): '[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x16_50e_sthv1_rgb Results: @@ -468,20 +436,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb/20201010_221240.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb/20201010_221240.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb/tsm_r50_1x1x16_50e_sthv1_rgb_20201010-17fa49f6.pth - gpu_mem(M): '10390' reference top1 acc (efficient/accurate): '[47.05 / 48.61](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[76.40 / 77.96](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 42856686 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r101_1x1x8_50e_sthv1_rgb Results: @@ -495,20 +462,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb/20201010_224055.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb/20201010_224055.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb/tsm_r101_1x1x8_50e_sthv1_rgb_20201010-43fedf2e.pth - gpu_mem(M): '9800' reference top1 acc (efficient/accurate): '[46.64 / 48.13](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[75.40 / 77.31](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 240 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 240 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x8_50e_sthv2_rgb Results: @@ -522,20 +488,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/20200912_140737.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/20200912_140737.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/tsm_r50_1x1x8_50e_sthv2_rgb_20200912-033c4ac6.pth - gpu_mem(M): '7069' reference top1 acc (efficient/accurate): '[57.98 / 60.69](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[84.57 / 86.28](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 256 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x8_50e_sthv2_rgb Results: @@ -549,20 +514,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/20210401_143656.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/20210401_143656.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/tsm_r50_256h_1x1x8_50e_sthv2_rgb_20210401-df97f3e1.pth - gpu_mem(M): '7069' reference top1 acc (efficient/accurate): '[xx / 61.2](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[xx / xx](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 240 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 240 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x16_50e_sthv2_rgb Results: @@ -576,20 +540,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/20201010_224215.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/20201010_224215.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/tsm_r50_1x1x16_50e_sthv2_rgb_20201010-16469c6f.pth - gpu_mem(M): '10400' reference top1 acc (efficient/accurate): '[58.90 / 60.98](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[85.29 / 86.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 6 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 6 + Pretrained: ImageNet + Resolution: height 256 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 256 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_1x1x16_50e_sthv2_rgb Results: @@ -603,20 +566,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/20210331_134458.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/20210331_134458.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/tsm_r50_256h_1x1x16_50e_sthv2_rgb_20210331-0a45549c.pth - gpu_mem(M): '10400' reference top1 acc (efficient/accurate): '[xx / 63.1](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[xx / xx](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb.py In Collection: TSM Metadata: Architecture: ResNet101 + Batch Size: 8 Epochs: 50 Parameters: 42856686 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 240 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 240 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r101_1x1x8_50e_sthv2_rgb Results: @@ -630,20 +592,19 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb/20201010_224100.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb/20201010_224100.log Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb/tsm_r101_1x1x8_50e_sthv2_rgb_20201010-98cdedb8.pth - gpu_mem(M): '9784' reference top1 acc (efficient/accurate): '[58.89 / 61.36](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' reference top5 acc (efficient/accurate): '[85.14 / 87.00](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)' - Config: configs/recognition/tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.py In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_mixup_1x1x8_50e_sthv1_rgb Results: @@ -663,13 +624,13 @@ Models: In Collection: TSM Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 8 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsm_r50_cutmix_1x1x8_50e_sthv1_rgb Results: @@ -685,3 +646,25 @@ Models: Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb-34934615.pth delta top1 acc (efficient/accurate): +0.34 / -0.24 delta top5 acc (efficient/accurate): +0.21 / +0.59 +- Config: configs/recognition/tsm/tsm_r50_1x1x8_50e_jester_rgb.py + In Collection: TSM + Metadata: + Architecture: ResNet50 + Batch Size: 8 + Epochs: 50 + Parameters: 23563355 + Pretrained: ImageNet + Resolution: height 100 + Training Data: Jester + Training Resources: 8 GPUs + Modality: RGB + Name: tsm_r50_1x1x8_50e_jester_rgb + Results: + - Dataset: Jester + Metrics: + - top1 acc (efficient): 96.5 + - top1 acc (accurate): 97.2 + Task: Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_jester_rgb/tsm_r50_1x1x8_50e_jester_rgb.json + Training Log: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_jester_rgb/tsm_r50_1x1x8_50e_jester_rgb.log + Weights: https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_jester_rgb/tsm_r50_1x1x8_50e_jester_rgb-c799267e.pth diff --git a/configs/recognition/tsn/metafile.yml b/configs/recognition/tsn/metafile.yml index 0c0a00aac1..d17f50e036 100644 --- a/configs/recognition/tsn/metafile.yml +++ b/configs/recognition/tsn/metafile.yml @@ -1,19 +1,17 @@ Collections: -- Metadata: - Training Data: null - Name: TSN +- Name: TSN README: configs/recognition/tsn/README.md Models: - Config: configs/recognition/tsn/tsn_r50_1x1x3_75e_ucf101_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 75 Parameters: 23714981 - Training BatchSize / GPU: 32 + Pretrained: ImageNet Training Data: UCF101 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x3_75e_ucf101_rgb Results: @@ -25,18 +23,16 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x3_75e_ucf101_rgb/tsn_r50_1x1x3_75e_ucf101_rgb_20201023.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x3_75e_ucf101_rgb/tsn_r50_1x1x3_75e_ucf101_rgb_20201023.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x3_75e_ucf101_rgb/tsn_r50_1x1x3_75e_ucf101_rgb_20201023-d85ab600.pth - config: '[tsn_r50_1x1x3_75e_ucf101_rgb](/configs/recognition/tsn/tsn_r50_1x1x3_75e_ucf101_rgb.py)' - gpu_mem(M): '8332' - Config: configs/recognition/tsn/tsn_r50_video_1x1x8_100e_diving48_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 100 Parameters: 23606384 - Training BatchSize / GPU: 8 + Pretrained: ImageNet Training Data: Diving48 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_video_1x1x8_100e_diving48_rgb Results: @@ -48,17 +44,16 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_diving48_rgb/20210426_014138.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_diving48_rgb/20210426_014138.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_diving48_rgb/tsn_r50_video_1x1x8_100e_diving48_rgb_20210426-6dde0185.pth - gpu_mem(M): '5699' - Config: configs/recognition/tsn/tsn_r50_video_1x1x16_100e_diving48_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 4 Epochs: 100 Parameters: 23606384 - Training BatchSize / GPU: 4 + Pretrained: ImageNet Training Data: Diving48 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_video_1x1x16_100e_diving48_rgb Results: @@ -70,17 +65,16 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x16_100e_diving48_rgb/20210426_014103.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x16_100e_diving48_rgb/20210426_014103.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x16_100e_diving48_rgb/tsn_r50_video_1x1x16_100e_diving48_rgb_20210426-63c5f2f7.pth - gpu_mem(M): '5705' - Config: configs/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_imagenet_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 50 Parameters: 23612531 - Training BatchSize / GPU: 32 + Pretrained: ImageNet Training Data: HMDB51 - gpus: 8 - pretrain: ImageNet + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x8_50e_hmdb51_imagenet_rgb Results: @@ -92,17 +86,16 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_imagenet_rgb/20201025_231108.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_imagenet_rgb/20201025_231108.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_imagenet_rgb/tsn_r50_1x1x8_50e_hmdb51_imagenet_rgb_20201123-ce6c27ed.pth - gpu_mem(M): '21535' - Config: configs/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 50 Parameters: 23612531 - Training BatchSize / GPU: 32 + Pretrained: Kinetics400 Training Data: HMDB51 - gpus: 8 - pretrain: Kinetics400 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x8_50e_hmdb51_kinetics400_rgb Results: @@ -114,16 +107,15 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_kinetics400_rgb/20201108_190805.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_kinetics400_rgb/20201108_190805.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_kinetics400_rgb/tsn_r50_1x1x8_50e_hmdb51_kinetics400_rgb_20201123-7f84701b.pth - gpu_mem(M): '21535' - Config: configs/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_mit_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 Epochs: 50 Parameters: 23612531 + Pretrained: Moments Training Data: HMDB51 - gpus: 8 - pretrain: Moments + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x8_50e_hmdb51_mit_rgb Results: @@ -135,18 +127,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_mit_rgb/20201112_170135.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_mit_rgb/20201112_170135.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_hmdb51_mit_rgb/tsn_r50_1x1x8_50e_hmdb51_mit_rgb_20201123-01526d41.pth - gpu_mem(M): '21535' - Config: configs/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x3_100e_kinetics400_rgb Results: @@ -158,19 +149,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb/20200614_063526.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb/20200614_063526.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb/tsn_r50_1x1x3_100e_kinetics400_rgb_20200614-e508be42.pth - gpu_mem(M): '8344' - inference_time(video/s): 4.3 (25x10 frames) - Config: configs/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x3_100e_kinetics400_rgb Results: @@ -182,19 +171,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_256p_1x1x3_100e_kinetics400_rgb/20200725_031325.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_256p_1x1x3_100e_kinetics400_rgb/20200725_031325.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_256p_1x1x3_100e_kinetics400_rgb/tsn_r50_256p_1x1x3_100e_kinetics400_rgb_20200725-22592236.pth - gpu_mem(M): '8343' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_dense_1x1x5_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 24 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 24 GPUs Modality: RGB Name: tsn_r50_dense_1x1x5_50e_kinetics400_rgb Results: @@ -206,19 +193,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_dense_1x1x5_100e_kinetics400_rgb/20200627_105310.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_dense_1x1x5_100e_kinetics400_rgb/20200627_105310.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_dense_1x1x5_100e_kinetics400_rgb/tsn_r50_dense_1x1x5_100e_kinetics400_rgb_20200627-a063165f.pth - gpu_mem(M): '7028' - inference_time(video/s): 12.7 (8x10 frames) - Config: configs/recognition/tsn/tsn_r50_320p_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: tsn_r50_320p_1x1x3_100e_kinetics400_rgb Results: @@ -230,19 +215,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x3_100e_kinetics400_rgb/tsn_r50_f3_kinetics400_shortedge_70.9_89.5.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x3_100e_kinetics400_rgb/tsn_r50_f3_kinetics400_shortedge_70.9_89.5.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x3_100e_kinetics400_rgb/tsn_r50_320p_1x1x3_100e_kinetics400_rgb_20200702-cc665e2a.pth - gpu_mem(M): '8344' - inference_time(video/s): 10.7 (25x3 frames) - Config: configs/recognition/tsn/tsn_r50_320p_1x1x3_110e_kinetics400_flow.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 110 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: Flow Name: tsn_r50_320p_1x1x3_110e_kinetics400_flow Results: @@ -254,19 +237,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x3_110e_kinetics400_flow/tsn_r50_f3_kinetics400_flow_shortedge_55.7_79.9.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x3_110e_kinetics400_flow/tsn_r50_f3_kinetics400_flow_shortedge_55.7_79.9.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x3_110e_kinetics400_flow/tsn_r50_320p_1x1x3_110e_kinetics400_flow_20200705-3036bab6.pth - gpu_mem(M): '8471' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x8_100e_kinetics400_rgb Results: @@ -278,19 +259,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_256p_1x1x8_100e_kinetics400_rgb/20200815_173413.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_256p_1x1x8_100e_kinetics400_rgb/20200815_173413.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_256p_1x1x8_100e_kinetics400_rgb/tsn_r50_256p_1x1x8_100e_kinetics400_rgb_20200817-883baf16.pth - gpu_mem(M): '8343' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 24 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 24 GPUs Modality: RGB Name: tsn_r50_320p_1x1x8_100e_kinetics400_rgb Results: @@ -302,19 +281,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_100e_kinetics400_rgb/tsn_r50_f8_kinetics400_shortedge_72.4_90.6.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_100e_kinetics400_rgb/tsn_r50_f8_kinetics400_shortedge_72.4_90.6.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_100e_kinetics400_rgb/tsn_r50_320p_1x1x8_100e_kinetics400_rgb_20200702-ef80e3d7.pth - gpu_mem(M): '8344' - inference_time(video/s): 11.1 (25x3 frames) - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_110e_kinetics400_flow.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 110 Parameters: 24327632 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 32 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 32 GPUs Modality: Flow Name: tsn_r50_320p_1x1x8_110e_kinetics400_flow Results: @@ -326,19 +303,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_110e_kinetics400_flow/tsn_r50_f8_kinetics400_flow_shortedge_57.8_81.0.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_110e_kinetics400_flow/tsn_r50_f8_kinetics400_flow_shortedge_57.8_81.0.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_110e_kinetics400_flow/tsn_r50_320p_1x1x8_110e_kinetics400_flow_20200705-1f39486b.pth - gpu_mem(M): '8473' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb Results: @@ -350,19 +325,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb_20201014.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb_20201014.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb/tsn_r50_video_320p_1x1x3_100e_kinetics400_rgb_20201014-5ae1ee79.pth - gpu_mem(M): '8343' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_dense_1x1x8_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: 340x256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_dense_1x1x8_100e_kinetics400_rgb Results: @@ -374,19 +347,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_dense_1x1x8_100e_kinetics400_rgb/20200606_003901.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_dense_1x1x8_100e_kinetics400_rgb/20200606_003901.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_dense_1x1x8_100e_kinetics400_rgb/tsn_r50_dense_1x1x8_100e_kinetics400_rgb_20200606-e925e6e3.pth - gpu_mem(M): '8344' - inference_time(video/s): 12.2 (8x10 frames) - Config: configs/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_video_1x1x8_100e_kinetics400_rgb Results: @@ -398,19 +369,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics400_rgb/tsn_r50_video_2d_1x1x8_100e_kinetics400_rgb.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics400_rgb/tsn_r50_video_2d_1x1x8_100e_kinetics400_rgb.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics400_rgb/tsn_r50_video_1x1x8_100e_kinetics400_rgb_20200702-568cde33.pth - gpu_mem(M): '21558' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_video_dense_1x1x8_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-400 - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_video_dense_1x1x8_100e_kinetics400_rgb Results: @@ -422,19 +391,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_dense_1x1x8_100e_kinetics400_rgb/tsn_r50_video_2d_1x1x8_dense_100e_kinetics400_rgb.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_dense_1x1x8_100e_kinetics400_rgb/tsn_r50_video_2d_1x1x8_dense_100e_kinetics400_rgb.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_dense_1x1x8_100e_kinetics400_rgb/tsn_r50_video_dense_1x1x8_100e_kinetics400_rgb_20200703-0f19175f.pth - gpu_mem(M): '21553' - inference_time(video/s): x - Config: configs/recognition/tsn/custom_backbones/tsn_rn101_32x4d_320p_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNeXt101-32x4d [[MMCls](https://github.com/open-mmlab/mmclassification/tree/master/configs/resnext)] + Batch Size: 16 Epochs: 100 Parameters: 42948304 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: tsn_rn101_32x4d_320p_1x1x3_100e_kinetics400_rgb Results: @@ -450,13 +417,13 @@ Models: In Collection: TSN Metadata: Architecture: ResNeXt101-32x4d [[TorchVision](https://github.com/pytorch/vision/)] + Batch Size: 12 Epochs: 100 Parameters: 27355600 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 320 Training Data: Kinetics-400 - gpus: 16 - pretrain: ImageNet - resolution: short-side 320 + Training Resources: 16 GPUs Modality: RGB Name: tsn_dense161_320p_1x1x3_100e_kinetics400_rgb Results: @@ -472,12 +439,12 @@ Models: In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: 340x256 Training Data: Kinetics-400 - pretrain: ImageNet - resolution: 340x256 Modality: RGB Name: tsn_omnisource_r50_1x1x3_100e_kinetics_rgb Results: @@ -487,18 +454,16 @@ Models: top5 acc: 91.0 Task: Action Recognition Weights: https://download.openmmlab.com/mmaction/recognition/tsn/omni/tsn_imagenet_pretrained_r50_omni_1x1x3_kinetics400_rgb_20200926-54192355.pth - gpu_mem(M): '8344' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: IG-1B + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: IG-1B - resolution: short-side 320 Modality: RGB Name: tsn_IG1B_pretrained_r50_1x1x3_100e_kinetics_rgb Results: @@ -508,18 +473,16 @@ Models: top5 acc: 90.4 Task: Action Recognition Weights: https://download.openmmlab.com/mmaction/recognition/tsn/omni/tsn_1G1B_pretrained_r50_without_omni_1x1x3_kinetics400_rgb_20200926-c133dd49.pth - gpu_mem(M): '8344' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 32 Epochs: 100 Parameters: 24327632 - Training BatchSize / GPU: 32 + Pretrained: IG-1B + Resolution: short-side 320 Training Data: Kinetics-400 - pretrain: IG-1B - resolution: short-side 320 Modality: RGB Name: tsn_IG1B_pretrained_omnisource_r50_1x1x3_100e_kinetics_rgb Results: @@ -529,19 +492,17 @@ Models: top5 acc: 91.9 Task: Action Recognition Weights: https://download.openmmlab.com/mmaction/recognition/tsn/omni/tsn_1G1B_pretrained_r50_omni_1x1x3_kinetics400_rgb_20200926-2863fed0.pth - gpu_mem(M): '8344' - inference_time(video/s): x - Config: configs/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Parameters: 24737432 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-600 - gpus: 16 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 16 GPUs Modality: RGB Name: tsn_r50_video_1x1x8_100e_kinetics600_rgb Results: @@ -553,19 +514,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb/tsn_r50_video_1x1x8_100e_kinetics600_rgb_20201015.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb/tsn_r50_video_1x1x8_100e_kinetics600_rgb_20201015.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics600_rgb/tsn_r50_video_1x1x8_100e_kinetics600_rgb_20201015-4db3c461.pth - gpu_mem(M): '8344' - inference_time(video/s): 11.1 (25x3 frames) - Config: configs/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 12 Epochs: 100 Parameters: 24942332 - Training BatchSize / GPU: 12 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: Kinetics-700 - gpus: 16 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 16 GPUs Modality: RGB Name: tsn_r50_video_1x1x8_100e_kinetics700_rgb Results: @@ -577,19 +536,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb/tsn_r50_video_1x1x8_100e_kinetics700_rgb_20201015.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb/tsn_r50_video_1x1x8_100e_kinetics700_rgb_20201015.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_video_1x1x8_100e_kinetics700_rgb/tsn_r50_video_1x1x8_100e_kinetics700_rgb_20201015-e381a6c7.pth - gpu_mem(M): '8344' - inference_time(video/s): 11.1 (25x3 frames) - Config: configs/recognition/tsn/tsn_r50_1x1x8_50e_sthv1_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x8_50e_sthv1_rgb Results: @@ -601,18 +558,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_sthv1_rgb/tsn_r50_f8_sthv1_18.1_45.0.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_sthv1_rgb/tsn_sthv1.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_sthv1_rgb/tsn_r50_1x1x8_50e_sthv1_rgb_20200618-061b9195.pth - gpu_mem(M): '10978' - Config: configs/recognition/tsn/tsn_r50_1x1x16_50e_sthv1_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 4 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 4 + Pretrained: ImageNet + Resolution: height 100 Training Data: SthV1 - gpus: 8 - pretrain: ImageNet - resolution: height 100 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x16_50e_sthv1_rgb Results: @@ -624,18 +580,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x16_50e_sthv1_rgb/20200614_211932.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x16_50e_sthv1_rgb/20200614_211932.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x16_50e_sthv1_rgb/tsn_r50_1x1x16_50e_sthv1_rgb_20200614-7e2fe4f1.pth - gpu_mem(M): '5691' - Config: configs/recognition/tsn/tsn_r50_1x1x8_50e_sthv2_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: height 240 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 240 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x8_50e_sthv2_rgb Results: @@ -647,18 +602,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_sthv2_rgb/20200915_114139.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_sthv2_rgb/20200915_114139.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x8_50e_sthv2_rgb/tsn_r50_1x1x8_50e_sthv2_rgb_20200915-f3b381a5.pth - gpu_mem(M): '10966' - Config: configs/recognition/tsn/tsn_r50_1x1x16_50e_sthv2_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 4 Epochs: 50 Parameters: 23864558 - Training BatchSize / GPU: 4 + Pretrained: ImageNet + Resolution: height 240 Training Data: SthV2 - gpus: 8 - pretrain: ImageNet - resolution: height 240 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_1x1x16_50e_sthv2_rgb Results: @@ -670,18 +624,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x16_50e_sthv2_rgb/20200917_105855.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x16_50e_sthv2_rgb/20200917_105855.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x16_50e_sthv2_rgb/tsn_r50_1x1x16_50e_sthv2_rgb_20200917-80bc3611.pth - gpu_mem(M): '8337' - Config: configs/recognition/tsn/tsn_r50_1x1x6_100e_mit_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 16 Epochs: 100 Parameters: 24202643 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: MiT - gpus: 16 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 16 GPUs Modality: RGB Name: tsn_r50_1x1x6_100e_mit_rgb Results: @@ -693,18 +646,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x6_100e_mit_rgb/tsn_r50_f6_mit_26.8_51.6.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x6_100e_mit_rgb/tsn_mit.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_1x1x6_100e_mit_rgb/tsn_r50_1x1x6_100e_mit_rgb_20200618-d512ab1b.pth - gpu_mem(M): '8339' - Config: configs/recognition/tsn/tsn_r101_1x1x5_50e_mmit_rgb.py In Collection: TSN Metadata: Architecture: ResNet101 + Batch Size: 16 Epochs: 50 Parameters: 43141497 - Training BatchSize / GPU: 16 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: MMiT - gpus: 16 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 16 GPUs Modality: RGB Name: tsn_r101_1x1x5_50e_mmit_rgb Results: @@ -715,18 +667,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r101_1x1x5_50e_mmit_rgb/tsn_r101_f6_mmit_61.1.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r101_1x1x5_50e_mmit_rgb/tsn_mmit.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r101_1x1x5_50e_mmit_rgb/tsn_r101_1x1x5_50e_mmit_rgb_20200618-642f450d.pth - gpu_mem(M): '10467' - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_video_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23917832 - Training BatchSize / GPU: 8 + Pretrained: Kinetics400 + Resolution: short-side 320 Training Data: ActivityNet v1.3 - gpus: 8 - pretrain: Kinetics400 - resolution: short-side 320 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_320p_1x1x8_50e_activitynet_video_rgb Results: @@ -738,18 +689,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_video_rgb/20210228_223327.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_video_rgb/20210228_223327.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_video_rgb/tsn_r50_320p_1x1x8_50e_activitynet_video_rgb_20210301-7f8da0c6.pth - gpu_mem(M): '5692' - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_clip_rgb.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 50 Parameters: 23917832 - Training BatchSize / GPU: 8 + Pretrained: Kinetics400 + Resolution: short-side 320 Training Data: ActivityNet v1.3 - gpus: 8 - pretrain: Kinetics400 - resolution: short-side 320 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r50_320p_1x1x8_50e_activitynet_clip_rgb Results: @@ -761,18 +711,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_clip_rgb/20210217_181313.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_clip_rgb/20210217_181313.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_50e_activitynet_clip_rgb/tsn_r50_320p_1x1x8_50e_activitynet_clip_rgb_20210301-c0f04a7e.pth - gpu_mem(M): '5692' - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_video_flow.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 23939784 - Training BatchSize / GPU: 8 + Pretrained: Kinetics400 + Resolution: 340x256 Training Data: ActivityNet v1.3 - gpus: 16 - pretrain: Kinetics400 - resolution: 340x256 + Training Resources: 16 GPUs Modality: Flow Name: tsn_r50_320p_1x1x8_150e_activitynet_video_flow Results: @@ -784,18 +733,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_video_flow/tsn_r50_320p_1x1x8_150e_activitynet_video_flow_20200804.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_video_flow/tsn_r50_320p_1x1x8_150e_activitynet_video_flow_20200804.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_video_flow/tsn_r50_320p_1x1x8_150e_activitynet_video_flow_20200804-13313f52.pth - gpu_mem(M): '5780' - Config: configs/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow.py In Collection: TSN Metadata: Architecture: ResNet50 + Batch Size: 8 Epochs: 150 Parameters: 23939784 - Training BatchSize / GPU: 8 + Pretrained: Kinetics400 + Resolution: 340x256 Training Data: ActivityNet v1.3 - gpus: 16 - pretrain: Kinetics400 - resolution: 340x256 + Training Resources: 16 GPUs Modality: Flow Name: tsn_r50_320p_1x1x8_150e_activitynet_clip_flow Results: @@ -807,18 +755,17 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow_20200804.json Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow_20200804.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow/tsn_r50_320p_1x1x8_150e_activitynet_clip_flow_20200804-8622cf38.pth - gpu_mem(M): '5780' - Config: configs/recognition/tsn/hvu/tsn_r18_1x1x8_100e_hvu_action_rgb.py In Collection: TSN Metadata: Architecture: ResNet18 + Batch Size: 32 Epochs: 100 Parameters: 11555619 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: HVU - gpus: 16 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 16 GPUs Modality: RGB Name: tsn_r18_1x1x8_100e_hvu_action_rgb Results: @@ -834,13 +781,13 @@ Models: In Collection: TSN Metadata: Architecture: ResNet18 + Batch Size: 32 Epochs: 100 Parameters: 11303736 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: HVU - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r18_1x1x8_100e_hvu_scene_rgb Results: @@ -856,13 +803,13 @@ Models: In Collection: TSN Metadata: Architecture: ResNet18 + Batch Size: 32 Epochs: 100 Parameters: 12037326 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: HVU - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r18_1x1x8_100e_hvu_object_rgb Results: @@ -878,13 +825,13 @@ Models: In Collection: TSN Metadata: Architecture: ResNet18 + Batch Size: 32 Epochs: 100 Parameters: 11211909 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: HVU - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r18_1x1x8_100e_hvu_event_rgb Results: @@ -900,13 +847,13 @@ Models: In Collection: TSN Metadata: Architecture: ResNet18 + Batch Size: 32 Epochs: 100 Parameters: 12037326 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: HVU - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r18_1x1x8_100e_hvu_concept_rgb Results: @@ -922,13 +869,13 @@ Models: In Collection: TSN Metadata: Architecture: ResNet18 + Batch Size: 32 Epochs: 100 Parameters: 11236533 - Training BatchSize / GPU: 32 + Pretrained: ImageNet + Resolution: short-side 256 Training Data: HVU - gpus: 8 - pretrain: ImageNet - resolution: short-side 256 + Training Resources: 8 GPUs Modality: RGB Name: tsn_r18_1x1x8_100e_hvu_attribute_rgb Results: @@ -940,3 +887,24 @@ Models: Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/hvu/attribute/tsn_r18_1x1x8_100e_hvu_attribute_rgb_20201027.log Weights: https://download.openmmlab.com/mmaction/recognition/tsn/hvu/attribute/tsn_r18_1x1x8_100e_hvu_attribute_rgb_20201027-0b3b49d2.pth tag category: attribute +- Config: configs/recognition/tsn/custom_backbones/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb.py + In Collection: TSN + Metadata: + Architecture: Swin Transformer + Batch Size: 24 + Epochs: 100 + Parameters: 87153224 + Pretrained: ImageNet + Resolution: short-side 320 + Training Data: Kinetics400 + Training Resources: 8 GPUs + Name: tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb + Results: + - Dataset: Kinetics400 + Metrics: + - top1 acc: 77.51 + - top5 acc: 92.92 + Task: Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/recognition/tsn/custom_backbones/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb.json + Training Log: https://download.openmmlab.com/mmaction/recognition/tsn/custom_backbones/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb.log + Weights: https://download.openmmlab.com/mmaction/recognition/tsn/custom_backbones/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb/tsn_swin_transformer_video_320p_1x1x3_100e_kinetics400_rgb-805380f6.pth diff --git a/configs/recognition/x3d/metafile.yml b/configs/recognition/x3d/metafile.yml index 2e4bde7889..3f7e9393d1 100644 --- a/configs/recognition/x3d/metafile.yml +++ b/configs/recognition/x3d/metafile.yml @@ -1,17 +1,15 @@ Collections: -- Metadata: - Training Data: null - Name: X3D +- Name: X3D README: configs/recognition/x3d/README.md Models: - Config: configs/recognition/x3d/x3d_s_13x6x1_facebook_kinetics400_rgb.py In Collection: X3D Metadata: Architecture: X3D_S + Batch Size: 1 Parameters: 3794322 - Training BatchSize / GPU: 1 + Resolution: short-side 320 Training Data: Kinetics-400 - resolution: short-side 320 Modality: RGB Name: x3d_s_13x6x1_facebook_kinetics400_rgb Results: @@ -27,10 +25,10 @@ Models: In Collection: X3D Metadata: Architecture: X3D_M + Batch Size: 1 Parameters: 3794322 - Training BatchSize / GPU: 1 + Resolution: short-side 320 Training Data: Kinetics-400 - resolution: short-side 320 Modality: RGB Name: x3d_m_16x5x1_facebook_kinetics400_rgb Results: diff --git a/configs/recognition_audio/resnet/metafile.yml b/configs/recognition_audio/resnet/metafile.yml index 3d139fd062..a13b5fc594 100644 --- a/configs/recognition_audio/resnet/metafile.yml +++ b/configs/recognition_audio/resnet/metafile.yml @@ -1,17 +1,15 @@ Collections: -- Metadata: - Training Data: null - Name: Audio +- Name: Audio README: configs/recognition_audio/resnet/README.md Models: - Config: configs/recognition_audio/resnet/tsn_r18_64x1x1_100e_kinetics400_audio_feature.py In Collection: Audio Metadata: Architecture: ResNet18 + Pretrained: None Training Data: Kinetics-400 - gpus: 8 + Training Resources: 8 GPUs n_fft: '1024' - pretrain: None Modality: Audio Name: tsn_r18_64x1x1_100e_kinetics400_audio_feature Results: @@ -27,4 +25,3 @@ Models: Training Json Log: https://download.openmmlab.com/mmaction/recognition/audio_recognition/tsn_r18_64x1x1_100e_kinetics400_audio_feature/20201010_144630.log.json Training Log: https://download.openmmlab.com/mmaction/recognition/audio_recognition/tsn_r18_64x1x1_100e_kinetics400_audio_feature/20201010_144630.log Weights: https://download.openmmlab.com/mmaction/recognition/audio_recognition/tsn_r18_64x1x1_100e_kinetics400_audio_feature/tsn_r18_64x1x1_100e_kinetics400_audio_feature_20201012-bf34df6c.pth - gpu_mem(M): '1897' diff --git a/configs/skeleton/posec3d/metafile.yml b/configs/skeleton/posec3d/metafile.yml new file mode 100644 index 0000000000..24c6e30c70 --- /dev/null +++ b/configs/skeleton/posec3d/metafile.yml @@ -0,0 +1,118 @@ +Collections: +- Name: PoseC3D + README: configs/skeleton/posec3d/README.md +Models: +- Config: configs/skeleton/posec3d/slowonly_r50_u48_240e_gym_keypoint.py + In Collection: PoseC3D + Metadata: + Architecture: SlowOnly-R50 + Batch Size: 16 + Epochs: 240 + Parameters: 2044867 + Training Data: FineGYM + Training Resources: 16 GPUs + pseudo heatmap: keypoint + Name: slowonly_r50_u48_240e_gym_keypoint + Results: + - Dataset: FineGYM + Metrics: + - Mean Top-1: 93.7 + Task: Skeleton-based Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_gym_keypoint/slowonly_r50_u48_240e_gym_keypoint.json + Training Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_gym_keypoint/slowonly_r50_u48_240e_gym_keypoint.log + Weights: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_gym_keypoint/slowonly_r50_u48_240e_gym_keypoint-b07a98a0.pth +- Config: configs/skeleton/posec3d/slowonly_r50_u48_240e_gym_limb.py + In Collection: PoseC3D + Metadata: + Architecture: SlowOnly-R50 + Batch Size: 16 + Epochs: 240 + Parameters: 2044867 + Training Data: FineGYM + Training Resources: 16 GPUs + pseudo heatmap: limb + Name: slowonly_r50_u48_240e_gym_limb + Results: + - Dataset: FineGYM + Metrics: + - Mean Top-1: 94.0 + Task: Skeleton-based Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_gym_limb/slowonly_r50_u48_240e_gym_limb.json + Training Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_gym_limb/slowonly_r50_u48_240e_gym_limb.log + Weights: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_gym_limb/slowonly_r50_u48_240e_gym_limb-c0d7b482.pth +- Config: configs/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_keypoint.py + In Collection: PoseC3D + Metadata: + Architecture: SlowOnly-R50 + Batch Size: 16 + Epochs: 240 + Parameters: 2024860 + Training Data: NTU60-XSub + Training Resources: 16 GPUs + pseudo heatmap: keypoint + Name: slowonly_r50_u48_240e_ntu60_xsub_keypoint + Results: + - Dataset: NTU60-XSub + Metrics: + - Top-1: 93.7 + Task: Skeleton-based Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_keypoint/slowonly_r50_u48_240e_ntu60_xsub_keypoint.json + Training Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_keypoint/slowonly_r50_u48_240e_ntu60_xsub_keypoint.log + Weights: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_keypoint/slowonly_r50_u48_240e_ntu60_xsub_keypoint-f3adabf1.pth +- Config: configs/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_limb.py + In Collection: PoseC3D + Metadata: + Architecture: SlowOnly-R50 + Batch Size: 16 + Epochs: 240 + Parameters: 2024860 + Training Data: NTU60-XSub + Training Resources: 16 GPUs + pseudo heatmap: limb + Name: slowonly_r50_u48_240e_ntu60_xsub_limb + Results: + - Dataset: NTU60-XSub + Metrics: + - Top-1: 93.4 + Task: Skeleton-based Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_limb/slowonly_r50_u48_240e_ntu60_xsub_limb.json + Training Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_limb/slowonly_r50_u48_240e_ntu60_xsub_limb.log + Weights: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu60_xsub_limb/slowonly_r50_u48_240e_ntu60_xsub_limb-1d69006a.pth +- Config: configs/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_keypoint.py + In Collection: PoseC3D + Metadata: + Architecture: SlowOnly-R50 + Batch Size: 16 + Epochs: 240 + Parameters: 2055640 + Training Data: NTU120-XSub + Training Resources: 16 GPUs + pseudo heatmap: keypoint + Name: slowonly_r50_u48_240e_ntu120_xsub_keypoint + Results: + - Dataset: NTU120-XSub + Metrics: + - Top-1: 86.3 + Task: Skeleton-based Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_keypoint/slowonly_r50_u48_240e_ntu120_xsub_keypoint.json + Training Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_keypoint/slowonly_r50_u48_240e_ntu120_xsub_keypoint.log + Weights: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_keypoint/slowonly_r50_u48_240e_ntu120_xsub_keypoint-6736b03f.pth +- Config: configs/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_limb.py + In Collection: PoseC3D + Metadata: + Architecture: SlowOnly-R50 + Batch Size: 16 + Epochs: 240 + Parameters: 2055640 + Training Data: NTU120-XSub + Training Resources: 16 GPUs + pseudo heatmap: limb + Name: slowonly_r50_u48_240e_ntu120_xsub_limb + Results: + - Dataset: NTU120-XSub + Metrics: + - Top-1: 85.7 + Task: Skeleton-based Action Recognition + Training Json Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_limb/slowonly_r50_u48_240e_ntu120_xsub_limb.json + Training Log: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_limb/slowonly_r50_u48_240e_ntu120_xsub_limb.log + Weights: https://download.openmmlab.com/mmaction/skeleton/posec3d/slowonly_r50_u48_240e_ntu120_xsub_limb/slowonly_r50_u48_240e_ntu120_xsub_limb-803c2317.pth? diff --git a/model_zoo.yml b/model_zoo.yml index 21af3c4394..1336135c17 100644 --- a/model_zoo.yml +++ b/model_zoo.yml @@ -18,4 +18,6 @@ Import: - configs/recognition/trn/metafile.yml - configs/detection/ava/metafile.yml - configs/detection/lfb/metafile.yml +- configs/detection/acrn/metafile.yml - configs/recognition_audio/resnet/metafile.yml +- configs/skeleton/posec3d/metafile.yml