diff --git a/configs/recognition/arr_tsm/README.md b/configs/recognition/arr_tsm/README.md
new file mode 100644
index 0000000000..aae497f3c6
--- /dev/null
+++ b/configs/recognition/arr_tsm/README.md
@@ -0,0 +1,142 @@
+# TSM
+
+## Introduction
+
+[ALGORITHM]
+
+```BibTeX
+@inproceedings{lin2019tsm,
+  title={TSM: Temporal Shift Module for Efficient Video Understanding},
+  author={Lin, Ji and Gan, Chuang and Han, Song},
+  booktitle={Proceedings of the IEEE International Conference on Computer Vision},
+  year={2019}
+}
+```
+
+[BACKBONE]
+
+```BibTeX
+@article{NonLocal2018,
+  author =   {Xiaolong Wang and Ross Girshick and Abhinav Gupta and Kaiming He},
+  title =    {Non-local Neural Networks},
+  journal =  {CVPR},
+  year =     {2018}
+}
+```
+
+## Model Zoo
+
+### Kinetics-400
+
+|config | resolution | gpus | backbone | pretrain | top1 acc| top5 acc | reference top1 acc | reference top5 acc | inference_time(video/s) | gpu_mem(M)| ckpt | log| json|
+|:--|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|
+|[tsm_r50_1x1x8_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py) |340x256|8| ResNet50| ImageNet |70.24|89.56|[70.36](https://github.com/mit-han-lab/temporal-shift-module/blob/8d53d6fda40bea2f1b37a6095279c4b454d672bd/scripts/train_tsm_kinetics_rgb_8f.sh)|[89.49](https://github.com/mit-han-lab/temporal-shift-module/blob/8d53d6fda40bea2f1b37a6095279c4b454d672bd/scripts/train_tsm_kinetics_rgb_8f.sh)|74.0 (8x1 frames)| 7079 | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb/tsm_r50_1x1x8_50e_kinetics400_rgb_20200607-af7fb746.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb/20200607_211800.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb/20200607_211800.log.json)|
+|[tsm_r50_1x1x8_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py) |short-side 256|8| ResNet50| ImageNet |70.59|89.52|x|x|x|7079|[ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x8_50e_kinetics400_rgb/tsm_r50_256p_1x1x8_50e_kinetics400_rgb_20200726-020785e2.pth)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x8_50e_kinetics400_rgb/20200725_031623.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x8_50e_kinetics400_rgb/20200725_031623.log.json)|
+|[tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py](configs/recognition/tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py) |short-side 256|8| ResNet50| ImageNet |70.48|89.40|x|x|x|7076|[ckpt](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)|[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)|[json](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)|
+|[tsm_r50_video_1x1x8_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_video_1x1x8_50e_kinetics400_rgb.py) |short-side 256|8| ResNet50| ImageNet |70.25|89.66|[70.36](https://github.com/mit-han-lab/temporal-shift-module/blob/8d53d6fda40bea2f1b37a6095279c4b454d672bd/scripts/train_tsm_kinetics_rgb_8f.sh)|[89.49](https://github.com/mit-han-lab/temporal-shift-module/blob/8d53d6fda40bea2f1b37a6095279c4b454d672bd/scripts/train_tsm_kinetics_rgb_8f.sh)|74.0 (8x1 frames)| 7077 | [ckpt]( https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_100e_kinetics400_rgb/tsm_r50_video_1x1x8_100e_kinetics400_rgb_20200702-a77f4328.pth) | [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](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_video_1x1x8_100e_kinetics400_rgb/tsm_r50_video_2d_1x1x8_50e_kinetics400_rgb.log.json)|
+|[tsm_r50_dense_1x1x8_100e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py) |340x256|8x4| ResNet50 | ImageNet|72.9|90.44|[72.22](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#dense-sample)|[90.37](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#dense-sample)|11.5 (8x10 frames)| 7079 | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/tsm_r50_dense_1x1x8_100e_kinetics400_rgb_20200626-91a54551.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/20200626_213415.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/20200626_213415.log.json)|
+|[tsm_r50_dense_1x1x8_100e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py) |short-side 256|8| ResNet50 | ImageNet|73.38|91.02|x|x|x|7079|[ckpt](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)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_256p_1x1x8_100e_kinetics400_rgb/20200725_032043.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_dense_256p_1x1x8_100e_kinetics400_rgb/20200725_032043.log.json)|
+|[tsm_r50_1x1x16_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py) |340x256|8| ResNet50| ImageNet |72.09|90.37|[70.67](https://github.com/mit-han-lab/temporal-shift-module/blob/8d53d6fda40bea2f1b37a6095279c4b454d672bd/scripts/train_tsm_kinetics_rgb_16f.sh)|[89.98](https://github.com/mit-han-lab/temporal-shift-module/blob/8d53d6fda40bea2f1b37a6095279c4b454d672bd/scripts/train_tsm_kinetics_rgb_16f.sh)|47.0 (16x1 frames)| 10404  | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb/tsm_r50_340x256_1x1x16_50e_kinetics400_rgb_20201011-2f27f229.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb/20201011_205356.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb/20201011_205356.log.json)|
+|[tsm_r50_1x1x16_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py) |short-side 256|8x4| ResNet50| ImageNet |71.89|90.73|x|x|x|10398|[ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x16_50e_kinetics400_rgb/tsm_r50_256p_1x1x16_50e_kinetics400_rgb_20201010-85645c2a.pth)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x16_50e_kinetics400_rgb/20201010_224825.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_256p_1x1x16_50e_kinetics400_rgb/20201010_224825.log.json)|
+|[tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb.py)|short-side 320|8x4| ResNet50| ImageNet |72.03|90.25|71.81|90.36|x|8931|[ckpt](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)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200724_120023.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200724_120023.log.json)|
+|[tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb.py)|short-side 320|8x4| ResNet50| ImageNet |70.70|89.90|x|x|x|10125|[ckpt](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)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200815_210253.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/20200815_210253.log.json)|
+|[tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb](/configs/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb.py)|short-side 320|8x4|ResNet50| ImageNet |71.60|90.34|x|x|x|8358|[ckpt](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)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb/20200723_220442.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb/20200723_220442.log.json)|
+|[tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb](/configs/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb.py)|short-side 320|8|MobileNetV2| ImageNet |68.46|88.64|x|x|x|3385|[ckpt](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)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb/20210129_024936.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb/20210129_024936.log.json)|
+
+### Something-Something V1
+
+|config | resolution | gpus | backbone| pretrain | top1 acc (efficient/accurate)| top5 acc (efficient/accurate)| reference top1 acc (efficient/accurate)| reference top5 acc (efficient/accurate)| gpu_mem(M)  | ckpt | log| json|
+|:--|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|
+|[tsm_r50_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb.py) |height 100|8| ResNet50 | ImageNet| 45.58 / 47.70|75.02 / 76.12|[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 7077| [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb/tsm_r50_1x1x8_50e_sthv1_rgb_20210203-01dce462.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb/20210203_150227.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv1_rgb/20210203_150227.log.json)|
+|[tsm_r50_flip_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb.py) |height 100|8| ResNet50 | ImageNet| 47.10 / 48.51|76.02 / 77.56|[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 7077| [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb/tsm_r50_flip_1x1x8_50e_sthv1_rgb_20210203-12596f16.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb/20210203_145829.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb/20210203_145829.log.json)|
+|[tsm_r50_randaugment_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.py) |height 100|8| ResNet50 | ImageNet| 47.16 / 48.90|76.07 / 77.92|[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 7077| [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb_20210324-481268d9.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.json)|
+|[tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.py) |height 100|8| ResNet50 | ImageNet| 47.85 / 50.31|76.78 / 78.18|[45.50 / 47.33](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[74.34 / 76.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 7077| [ckpt](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) | [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)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.json)|
+|[tsm_r50_1x1x16_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb.py)|height 100|8| ResNet50 | ImageNet|47.62 / 49.28|76.63 / 77.82|[47.05 / 48.61](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[76.40 / 77.96](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|10390|[ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb/tsm_r50_1x1x16_50e_sthv1_rgb_20201010-17fa49f6.pth)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb/20201010_221240.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv1_rgb/20201010_221240.log.json)|
+|[tsm_r101_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb.py)|height 100|8| ResNet50 | ImageNet|45.72 / 48.43|74.67 / 76.72|[46.64 / 48.13](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[75.40 / 77.31](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|9800|[ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb/tsm_r101_1x1x8_50e_sthv1_rgb_20201010-43fedf2e.pth)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb/20201010_224055.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv1_rgb/20201010_224055.log.json)|
+
+### Something-Something V2
+
+|config | resolution | gpus | backbone | pretrain| top1 acc (efficient/accurate)| top5 acc (efficient/accurate)|  reference top1 acc (efficient/accurate)| reference top5 acc (efficient/accurate)| gpu_mem(M)  | ckpt | log| json|
+|:--|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|:--:|
+|[tsm_r50_1x1x8_50e_sthv2_rgb](/configs/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb.py) |height 240|8| ResNet50| ImageNet |57.86 / 61.12|84.67 / 86.26|[57.98 / 60.69](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[84.57 / 86.28](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 7069 | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/tsm_r50_1x1x8_50e_sthv2_rgb_20200912-033c4ac6.pth)|[log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/20200912_140737.log)|[json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x8_50e_sthv2_rgb/20200912_140737.log.json)|
+|[tsm_r50_1x1x16_50e_sthv2_rgb](/configs/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb.py) |height 240|8| ResNet50| ImageNet |59.93 / 62.04|86.10 / 87.35|[58.90 / 60.98](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[85.29 / 86.60](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 10400| [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/tsm_r50_1x1x16_50e_sthv2_rgb_20201010-16469c6f.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/20201010_224215.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_1x1x16_50e_sthv2_rgb/20201010_224215.log.json)|
+|[tsm_r101_1x1x8_50e_sthv2_rgb](/configs/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb.py) |height 240|8| ResNet101 | ImageNet|58.59 / 61.51|85.07 / 86.90|[58.89 / 61.36](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)|[85.14 / 87.00](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd#training)| 9784 | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb/tsm_r101_1x1x8_50e_sthv2_rgb_20201010-98cdedb8.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb/20201010_224100.log)| [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r101_1x1x8_50e_sthv2_rgb/20201010_224100.log.json)|
+
+### MixUp & CutMix on Something-Something V1
+
+| config                                                       | resolution | gpus | backbone | pretrain | top1 acc (efficient/accurate) | top5 acc (efficient/accurate) | delta top1 acc (efficient/accurate) | delta top5 acc (efficient/accurate) |                             ckpt                             |                             log                              |                             json                             |
+| :----------------------------------------------------------- | :--------: | :--: | :------: | :------: | :---------------------------: | :---------------------------: | :---------------------------------: | :---------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
+| [tsm_r50_mixup_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.py) | height 100 |  8   | ResNet50 | ImageNet |         46.35 / 48.49         |         75.07 / 76.88         |            +0.77 / +0.79            |            +0.05 / +0.70            | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb/tsm_r50_mixup_1x1x8_50e_sthv1_rgb-9eca48e5.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.log) | [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.json) |
+| [tsm_r50_cutmix_1x1x8_50e_sthv1_rgb](/configs/recognition/tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb.py) | height 100 |  8   | ResNet50 | ImageNet |         45.92 / 47.46         |         75.23 / 76.71         |            +0.34 / -0.24            |            +0.21 / +0.59            | [ckpt](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb-34934615.pth) | [log](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb.log) | [json](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb.json) |
+
+Notes:
+
+1. The **gpus** indicates the number of gpu we used to get the checkpoint. It is noteworthy that the configs we provide are used for 8 gpus as default.
+   According to the [Linear Scaling Rule](https://arxiv.org/abs/1706.02677), you may set the learning rate proportional to the batch size if you use different GPUs or videos per GPU,
+   e.g., lr=0.01 for 4 GPUs x 2 video/gpu and lr=0.08 for 16 GPUs x 4 video/gpu.
+2. The **inference_time** is got by this [benchmark script](/tools/analysis/benchmark.py), where we use the sampling frames strategy of the test setting and only care about the model inference time,
+   not including the IO time and pre-processing time. For each setting, we use 1 gpu and set batch size (videos per gpu) to 1 to calculate the inference time.
+3. The values in columns named after "reference" are the results got by training on the original repo, using the same model settings. The checkpoints for reference repo can be downloaded [here](https://download.openmmlab.com/mmaction/recognition/tsm/tsm_reference_ckpt.rar).
+4. There are two kinds of test settings for Something-Something dataset, efficient setting (center crop x 1 clip) and accurate setting (Three crop x 2 clip), which is referred from the [original repo](https://github.com/mit-han-lab/temporal-shift-module/tree/8d53d6fda40bea2f1b37a6095279c4b454d672bd).
+   We use efficient setting as default provided in config files, and it can be changed to accurate setting by
+
+```python
+...
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,   # `num_clips = 8` when using 8 segments
+        twice_sample=True,    # set `twice_sample=True` for twice sample in accurate setting
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    # dict(type='CenterCrop', crop_size=224), it is used for efficient setting
+    dict(type='ThreeCrop', crop_size=256),  # it is used for accurate setting
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+```
+
+For more details on data preparation, you can refer to Kinetics400, Something-Something V1 and Something-Something V2 in [Data Preparation](/docs/data_preparation.md).
+
+5. When applying Mixup and CutMix, we use the hyper parameter `alpha=0.2`.
+
+## Train
+
+You can use the following command to train a model.
+
+```shell
+python tools/train.py ${CONFIG_FILE} [optional arguments]
+```
+
+Example: train TSM model on Kinetics-400 dataset in a deterministic option with periodic validation.
+
+```shell
+python tools/train.py configs/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py \
+    --work-dir work_dirs/tsm_r50_1x1x8_100e_kinetics400_rgb \
+    --validate --seed 0 --deterministic
+```
+
+For more details, you can refer to **Training setting** part in [getting_started](/docs/getting_started.md#training-setting).
+
+## Test
+
+You can use the following command to test a model.
+
+```shell
+python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [optional arguments]
+```
+
+Example: test TSM model on Kinetics-400 dataset and dump the result to a json file.
+
+```shell
+python tools/test.py configs/recognition/tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py \
+    checkpoints/SOME_CHECKPOINT.pth --eval top_k_accuracy mean_class_accuracy \
+    --out result.json
+```
+
+For more details, you can refer to **Test a dataset** part in [getting_started](/docs/getting_started.md#test-a-dataset).
diff --git a/configs/recognition/arr_tsm/tsm_mobilenetv2_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_mobilenetv2_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..647cb609ac
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_mobilenetv2_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,92 @@
+_base_ = [
+    '../../_base_/models/tsm_mobilenet_v2.py',
+    '../../_base_/schedules/sgd_tsm_mobilenet_v2_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='ThreeCrop', crop_size=256),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.01,  # this lr is used for 8 gpus
+)
+
+# runtime settings
+checkpoint_config = dict(interval=1)
+work_dir = './work_dirs/tsm_mobilenetv2_dense_1x1x8_100e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_mobilenetv2_video_dense_1x1x8_100e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_mobilenetv2_video_dense_1x1x8_100e_kinetics400_rgb.py
new file mode 100644
index 0000000000..e0a3c4873b
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_mobilenetv2_video_dense_1x1x8_100e_kinetics400_rgb.py
@@ -0,0 +1,97 @@
+_base_ = [
+    '../../_base_/models/tsm_mobilenet_v2.py',
+    '../../_base_/schedules/sgd_tsm_mobilenet_v2_100e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# dataset settings
+dataset_type = 'VideoDataset'
+data_root = 'data/kinetics400/videos_train'
+data_root_val = 'data/kinetics400/videos_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_videos.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_videos.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_videos.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='DecordInit'),
+    dict(type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(type='DecordInit'),
+    dict(
+        type='DenseSampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(type='DecordInit'),
+    dict(
+        type='DenseSampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='ThreeCrop', crop_size=256),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.01,  # this lr is used for 8 gpus
+)
+
+# runtime settings
+checkpoint_config = dict(interval=5)
+work_dir = './work_dirs/tsm_mobilenetv2_dense_video_1x1x8_100e_kinetics400_rgb/'  # noqa
diff --git a/configs/recognition/arr_tsm/tsm_mobilenetv2_video_inference_dense_1x1x8_100e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_mobilenetv2_video_inference_dense_1x1x8_100e_kinetics400_rgb.py
new file mode 100644
index 0000000000..a66c772a9b
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_mobilenetv2_video_inference_dense_1x1x8_100e_kinetics400_rgb.py
@@ -0,0 +1,34 @@
+_base_ = ['../../_base_/models/tsm_mobilenet_v2.py']
+
+# dataset settings
+dataset_type = 'VideoDataset'
+data_root_val = 'data/kinetics400/videos_val'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_videos.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+test_pipeline = [
+    dict(type='DecordInit'),
+    dict(
+        type='DenseSampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+
+data = dict(
+    videos_per_gpu=4,
+    workers_per_gpu=4,
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
diff --git a/configs/recognition/arr_tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..a85104da17
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_nl_dot_product_r50_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,97 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(
+    backbone=dict(
+        non_local=((0, 0, 0), (1, 0, 1, 0), (1, 0, 1, 0, 1, 0), (0, 0, 0)),
+        non_local_cfg=dict(
+            sub_sample=True,
+            use_scale=False,
+            norm_cfg=dict(type='BN3d', requires_grad=True),
+            mode='dot_product')))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# runtime settings
+work_dir = './work_dirs/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..c501dcd6d0
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,97 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(
+    backbone=dict(
+        non_local=((0, 0, 0), (1, 0, 1, 0), (1, 0, 1, 0, 1, 0), (0, 0, 0)),
+        non_local_cfg=dict(
+            sub_sample=True,
+            use_scale=False,
+            norm_cfg=dict(type='BN3d', requires_grad=True),
+            mode='embedded_gaussian')))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# runtime settings
+work_dir = './work_dirs/tsm_nl_embedded_gaussian_r50_1x1x8_50e_kinetics400_rgb/'  # noqa: E501
diff --git a/configs/recognition/arr_tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..8713150fda
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,97 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(
+    backbone=dict(
+        non_local=((0, 0, 0), (1, 0, 1, 0), (1, 0, 1, 0, 1, 0), (0, 0, 0)),
+        non_local_cfg=dict(
+            sub_sample=True,
+            use_scale=False,
+            norm_cfg=dict(type='BN3d', requires_grad=True),
+            mode='gaussian')))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# runtime settings
+work_dir = './work_dirs/tsm_nl_gaussian_r50_1x1x8_50e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r101_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r101_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..02c43a3808
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r101_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,7 @@
+_base_ = ['./tsm_r50_1x1x8_50e_sthv1_rgb.py']
+
+# model settings
+model = dict(backbone=dict(pretrained='torchvision://resnet101', depth=101))
+
+# runtime settings
+work_dir = './work_dirs/tsm_r101_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r101_1x1x8_50e_sthv2_rgb.py b/configs/recognition/arr_tsm/tsm_r101_1x1x8_50e_sthv2_rgb.py
new file mode 100644
index 0000000000..a4c5ce7d41
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r101_1x1x8_50e_sthv2_rgb.py
@@ -0,0 +1,88 @@
+_base_ = ['./tsm_r50_1x1x8_50e_sthv2_rgb.py']
+
+# model settings
+model = dict(backbone=dict(pretrained='torchvision://resnet101', depth=101))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv2/rawframes'
+data_root_val = 'data/sthv2/rawframes'
+ann_file_train = 'data/sthv2/sthv2_train_list_rawframes.txt'
+ann_file_val = 'data/sthv2/sthv2_val_list_rawframes.txt'
+ann_file_test = 'data/sthv2/sthv2_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.01,  # this lr is used for 8 gpus
+)
+# runtime settings
+work_dir = './work_dirs/tsm_r101_1x1x8_50e_sthv2_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..c53bee6c09
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_kinetics400_rgb.py
@@ -0,0 +1,96 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(backbone=dict(num_segments=16), cls_head=dict(num_segments=16))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=16),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=6,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.0075,  # this lr is used for 8 gpus
+)
+
+# runtime settings
+checkpoint_config = dict(interval=5)
+work_dir = './work_dirs/tsm_r50_1x1x16_50e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..590a8d1ac9
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_sthv1_rgb.py
@@ -0,0 +1,97 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(
+    backbone=dict(num_segments=16),
+    cls_head=dict(num_classes=174, num_segments=16))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=16),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=6,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.0075,  # this lr is used for 8 gpus
+    weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_1x1x16_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_sthv2_rgb.py b/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_sthv2_rgb.py
new file mode 100644
index 0000000000..04bd982d77
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_1x1x16_50e_sthv2_rgb.py
@@ -0,0 +1,94 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(
+    backbone=dict(num_segments=16),
+    cls_head=dict(num_classes=174, num_segments=16))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv2/rawframes'
+data_root_val = 'data/sthv2/rawframes'
+ann_file_train = 'data/sthv2/sthv2_train_list_rawframes.txt'
+ann_file_val = 'data/sthv2/sthv2_val_list_rawframes.txt'
+ann_file_test = 'data/sthv2/sthv2_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=16),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=16,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=6,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.0075,  # this lr is used for 8 gpus
+    weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_1x1x16_50e_sthv2_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_household_rgb.py b/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_household_rgb.py
new file mode 100644
index 0000000000..821d717738
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_household_rgb.py
@@ -0,0 +1,92 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(cls_head=dict(num_classes=174))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/household/rawframes'
+data_root_val = 'data/household/rawframes'
+ann_file_train = 'data/household/household_train_list_rawframes.txt'
+ann_file_val = 'data/household/household_val_list_rawframes.txt'
+ann_file_test = 'data/household/household_test_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=6,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.0075,  # this lr is used for 8 gpus
+    weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_1x1x8_50e_household_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..ec8ef9678e
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,88 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# runtime settings
+checkpoint_config = dict(interval=5)
+work_dir = './work_dirs/tsm_r50_1x1x8_100e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..4967fa23ac
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,93 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(cls_head=dict(num_classes=174))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..abf672adc2
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,114 @@
+_base_ = [
+    '../../_base_/schedules/sgd_tsm_50e.py', '../../_base_/default_runtime.py'
+]
+
+# model settings
+# model settings# model settings
+model = dict(
+    type='Recognizer2D',
+    backbone=dict(
+        type='ResNetTSM',
+        pretrained='torchvision://resnet50',
+        depth=50,
+        norm_eval=False,
+        shift_div=8),
+    cls_head=dict(
+        type='TSMHead',
+        num_classes=174,
+        in_channels=2048,
+        spatial_type='avg',
+        consensus=dict(type='AvgConsensus', dim=1),
+        dropout_ratio=0.5,
+        init_std=0.001,
+        is_shift=True),
+    # model training and testing settings
+    train_cfg=dict(
+        blending=dict(type='CutmixBlending', num_classes=174, alpha=.2)),
+    test_cfg=dict(average_clips='prob'))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        twice_sample=True,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='ThreeCrop', crop_size=256),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_cutmix_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py
new file mode 100644
index 0000000000..5f81caa280
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_dense_1x1x8_100e_kinetics400_rgb.py
@@ -0,0 +1,88 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_100e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='DenseSampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='DenseSampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    val_dataloader=dict(videos_per_gpu=4),
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_dense_1x1x8_100e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..00f40cbd58
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_flip_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,97 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(cls_head=dict(num_classes=174))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+
+sthv1_flip_label_map = {2: 4, 4: 2, 30: 41, 41: 30, 52: 66, 66: 52}
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5, flip_label_map=sthv1_flip_label_map),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_flip_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..d03ba632b4
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,98 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(cls_head=dict(num_classes=174))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+
+sthv1_flip_label_map = {2: 4, 4: 2, 30: 41, 41: 30, 52: 66, 66: 52}
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5, flip_label_map=sthv1_flip_label_map),
+    dict(type='Imgaug', transforms='default'),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_flip_randaugment_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..803d793223
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_gpu_normalize_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,94 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+module_hooks = [
+    dict(
+        type='GPUNormalize',
+        hooked_module='backbone',
+        hook_pos='forward_pre',
+        input_format='NCHW',
+        mean=[123.675, 116.28, 103.53],
+        std=[58.395, 57.12, 57.375])
+]
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/kinetics400/rawframes_train'
+data_root_val = 'data/kinetics400/rawframes_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_rawframes.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_rawframes.txt'
+
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# runtime settings
+checkpoint_config = dict(interval=5)
+work_dir = './work_dirs/tsm_r50_gpu_normalize_1x1x8_100e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..73d6321081
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_mixup_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,113 @@
+_base_ = [
+    '../../_base_/schedules/sgd_tsm_50e.py', '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(
+    type='Recognizer2D',
+    backbone=dict(
+        type='ResNetTSM',
+        pretrained='torchvision://resnet50',
+        depth=50,
+        norm_eval=False,
+        shift_div=8),
+    cls_head=dict(
+        type='TSMHead',
+        num_classes=174,
+        in_channels=2048,
+        spatial_type='avg',
+        consensus=dict(type='AvgConsensus', dim=1),
+        dropout_ratio=0.5,
+        init_std=0.001,
+        is_shift=True),
+    # model training and testing settings
+    train_cfg=dict(
+        blending=dict(type='MixupBlending', num_classes=174, alpha=.2)),
+    test_cfg=dict(average_clips='prob'))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        twice_sample=True,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='ThreeCrop', crop_size=256),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_mixup_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.py b/configs/recognition/arr_tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.py
new file mode 100644
index 0000000000..448908c5b4
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb.py
@@ -0,0 +1,94 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# model settings
+model = dict(cls_head=dict(num_classes=174))
+
+# dataset settings
+dataset_type = 'RawframeDataset'
+data_root = 'data/sthv1/rawframes'
+data_root_val = 'data/sthv1/rawframes'
+ann_file_train = 'data/sthv1/sthv1_train_list_rawframes.txt'
+ann_file_val = 'data/sthv1/sthv1_val_list_rawframes.txt'
+ann_file_test = 'data/sthv1/sthv1_val_list_rawframes.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Imgaug', transforms='default'),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='RawFrameDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        filename_tmpl='{:05}.jpg',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        filename_tmpl='{:05}.jpg',
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=2, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(weight_decay=0.0005)
+
+# runtime settings
+work_dir = './work_dirs/tsm_r50_randaugment_1x1x8_50e_sthv1_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_video_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_r50_video_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..0a172d0205
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_video_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,95 @@
+_base_ = [
+    '../../_base_/models/tsm_r50.py', '../../_base_/schedules/sgd_tsm_50e.py',
+    '../../_base_/default_runtime.py'
+]
+
+# dataset settings
+dataset_type = 'VideoDataset'
+data_root = 'data/kinetics400/videos_train'
+data_root_val = 'data/kinetics400/videos_val'
+ann_file_train = 'data/kinetics400/kinetics400_train_list_videos.txt'
+ann_file_val = 'data/kinetics400/kinetics400_val_list_videos.txt'
+ann_file_test = 'data/kinetics400/kinetics400_val_list_videos.txt'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+train_pipeline = [
+    dict(type='DecordInit'),
+    dict(type='SampleFrames', clip_len=1, frame_interval=1, num_clips=8),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(
+        type='MultiScaleCrop',
+        input_size=224,
+        scales=(1, 0.875, 0.75, 0.66),
+        random_crop=False,
+        max_wh_scale_gap=1,
+        num_fixed_crops=13),
+    dict(type='Resize', scale=(224, 224), keep_ratio=False),
+    dict(type='Flip', flip_ratio=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs', 'label'])
+]
+val_pipeline = [
+    dict(type='DecordInit'),
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+test_pipeline = [
+    dict(type='DecordInit'),
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Flip', flip_ratio=0),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs', 'label'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+data = dict(
+    videos_per_gpu=8,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=ann_file_train,
+        data_prefix=data_root,
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=ann_file_val,
+        data_prefix=data_root_val,
+        pipeline=val_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=ann_file_test,
+        data_prefix=data_root_val,
+        pipeline=test_pipeline))
+evaluation = dict(
+    interval=5, metrics=['top_k_accuracy', 'mean_class_accuracy'])
+
+# optimizer
+optimizer = dict(
+    lr=0.02,  # this lr is used for 8 gpus
+)
+# runtime settings
+checkpoint_config = dict(interval=5)
+work_dir = './work_dirs/tsm_r50_video_2d_1x1x8_50e_kinetics400_rgb/'
diff --git a/configs/recognition/arr_tsm/tsm_r50_video_inference_1x1x8_100e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_r50_video_inference_1x1x8_100e_kinetics400_rgb.py
new file mode 100644
index 0000000000..7c355ade2f
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_r50_video_inference_1x1x8_100e_kinetics400_rgb.py
@@ -0,0 +1,31 @@
+_base_ = ['../../_base_/models/tsm_r50.py']
+
+# dataset settings
+dataset_type = 'VideoDataset'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_bgr=False)
+test_pipeline = [
+    dict(type='DecordInit', num_threads=1),
+    dict(
+        type='SampleFrames',
+        clip_len=1,
+        frame_interval=1,
+        num_clips=8,
+        test_mode=True),
+    dict(type='DecordDecode'),
+    dict(type='Resize', scale=(-1, 256)),
+    dict(type='CenterCrop', crop_size=224),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='FormatShape', input_format='NCHW'),
+    dict(type='Collect', keys=['imgs'], meta_keys=[]),
+    dict(type='ToTensor', keys=['imgs'])
+]
+
+data = dict(
+    videos_per_gpu=1,
+    workers_per_gpu=2,
+    test=dict(
+        type=dataset_type,
+        ann_file=None,
+        data_prefix=None,
+        pipeline=test_pipeline))
diff --git a/configs/recognition/arr_tsm/tsm_temporal_pool_r50_1x1x8_50e_kinetics400_rgb.py b/configs/recognition/arr_tsm/tsm_temporal_pool_r50_1x1x8_50e_kinetics400_rgb.py
new file mode 100644
index 0000000000..2984d37968
--- /dev/null
+++ b/configs/recognition/arr_tsm/tsm_temporal_pool_r50_1x1x8_50e_kinetics400_rgb.py
@@ -0,0 +1,8 @@
+_base_ = ['./tsm_r50_1x1x8_50e_kinetics400_rgb.py']
+
+# model settings
+model = dict(
+    backbone=dict(temporal_pool=True), cls_head=dict(temporal_pool=True))
+
+# runtime settings
+work_dir = './work_dirs/tsm_temporal_pool_r50_1x1x8_100e_kinetics400_rgb/'
diff --git a/docker/Dockerfile b/docker/Dockerfile
index 7a47d42e01..f8c89a472b 100644
--- a/docker/Dockerfile
+++ b/docker/Dockerfile
@@ -12,12 +12,12 @@ RUN apt-get update && apt-get install -y git ninja-build libglib2.0-0 libsm6 lib
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/*
 
+# Inserted to use caches
+ENV updated-on 29-March-21
+
 # Install mmcv-full
 RUN pip install mmcv-full==latest+torch1.6.0+cu101 -f https://download.openmmlab.com/mmcv/dist/index.html
 
-# Inserted to use caches
-ENV updated-on 26-March-21
-
 # Install MMAction2
 RUN conda clean --all
 RUN git clone https://github.com/open-mmlab/mmaction2.git /mmaction2
diff --git a/docker/run.sh b/docker/run.sh
index 139601e2ff..bb61fb66df 100755
--- a/docker/run.sh
+++ b/docker/run.sh
@@ -5,6 +5,10 @@ docker run --rm \
        --privileged \
        --volume="/dev:/dev" \
        --volume="/mnt/hdd/video/household:/mmaction2/data/household" \
+       --volume="/home/ubuntu18/Codes/actionrecognition/mmaction2/tools/data:/mmaction2/tools/data" \
+       --volume="/home/ubuntu18/Codes/actionrecognition/mmaction2/configs/recognition/arr_tsm:/mmaction2/configs/recognition/arr_tsm" \
+       --volume="/home/ubuntu18/Codes/actionrecognition/mmaction2/work_dirs:/mmaction2/work_dirs" \
+       --volume="/home/ubuntu18/Codes/actionrecognition/mmaction2/mmaction/datasets:/mmaction2/mmaction/datasets" \
        --runtime=nvidia \
        --device /dev/snd:/dev/snd \
        -e DISPLAY=$DISPLAY \
diff --git a/mmaction/datasets/rawframe_dataset.py b/mmaction/datasets/rawframe_dataset.py
index 6c632851f2..be6fd83a88 100644
--- a/mmaction/datasets/rawframe_dataset.py
+++ b/mmaction/datasets/rawframe_dataset.py
@@ -93,7 +93,7 @@ def __init__(self,
                  with_offset=False,
                  multi_class=False,
                  num_classes=None,
-                 start_index=1,
+                 start_index=0,#1
                  modality='RGB',
                  sample_by_class=False,
                  power=None):
diff --git a/tools/data/build_file_list.py b/tools/data/build_file_list.py
index b0c22a074b..0fa5121f51 100644
--- a/tools/data/build_file_list.py
+++ b/tools/data/build_file_list.py
@@ -155,8 +155,9 @@ def build_list(split):
         return rgb_list, flow_list
 
     train_rgb_list, train_flow_list = build_list(splits[0])
-    test_rgb_list, test_flow_list = build_list(splits[1])
-    return (train_rgb_list, test_rgb_list), (train_flow_list, test_flow_list)
+    valid_rgb_list, valid_flow_list = build_list(splits[1])
+    test_rgb_list, test_flow_list = build_list(splits[2])
+    return (train_rgb_list, valid_rgb_list, test_rgb_list), (train_flow_list, valid_flow_list, test_flow_list)
 
 
 def main():
diff --git a/tools/data/household/generate_rawframes_filelist.sh b/tools/data/household/generate_rawframes_filelist.sh
index 00c8d913fc..3b570231e7 100644
--- a/tools/data/household/generate_rawframes_filelist.sh
+++ b/tools/data/household/generate_rawframes_filelist.sh
@@ -3,6 +3,8 @@
 cd ../../../
 PYTHONPATH=. python tools/data/build_file_list.py household data/household/rawframes/ --num-split 1 --level 2 --subset train --format rawframes --shuffle
 PYTHONPATH=. python tools/data/build_file_list.py household data/household/rawframes/ --num-split 1 --level 2 --subset val --format rawframes --shuffle
+PYTHONPATH=. python tools/data/build_file_list.py household data/household/rawframes/ --num-split 1 --level 2 --subset test --format rawframes --shuffle
+
 echo "Filelist for rawframes generated."
 
 cd tools/data/household/
diff --git a/tools/data/household/generate_videos_filelist.sh b/tools/data/household/generate_videos_filelist.sh
index ee5e56860f..e50bb4dec0 100644
--- a/tools/data/household/generate_videos_filelist.sh
+++ b/tools/data/household/generate_videos_filelist.sh
@@ -1,8 +1,9 @@
 #!/usr/bin/env bash
 
 cd ../../../
-PYTHONPATH=. python tools/data/build_file_list.py household data/household/videos/ --num-split 1 --level 1 --subset train --format videos --shuffle
-PYTHONPATH=. python tools/data/build_file_list.py household data/household/videos/ --num-split 1 --level 1 --subset val --format videos --shuffle
+PYTHONPATH=. python tools/data/build_file_list.py household data/household/videos/ --num-split 1 --level 2 --subset train --format videos --shuffle
+PYTHONPATH=. python tools/data/build_file_list.py household data/household/videos/ --num-split 1 --level 2 --subset val --format videos --shuffle
+PYTHONPATH=. python tools/data/build_file_list.py household data/household/videos/ --num-split 1 --level 2 --subset test --format videos --shuffle
 echo "Filelist for videos generated."
 
 cd tools/data/household/
diff --git a/tools/data/parse_file_list.py b/tools/data/parse_file_list.py
index a937d5cae9..d9bc292fe6 100644
--- a/tools/data/parse_file_list.py
+++ b/tools/data/parse_file_list.py
@@ -290,11 +290,11 @@ def parse_household_splits(level):
     """
     # Read the annotations
     # yapf: disable
-    class_index_file = 'data/sthv2/annotations/shousehold-labels.json'  # noqa
+    class_index_file = 'data/household/annotations/household-labels.json'  # noqa
     # yapf: enable
-    train_file = 'data/sthv2/annotations/household-train.json'
-    val_file = 'data/sthv2/annotations/household-validation.json'
-    test_file = 'data/sthv2/annotations/household-test.json'
+    train_file = 'data/household/annotations/household-train.json'
+    val_file = 'data/household/annotations/household-validation.json'
+    test_file = 'data/household/annotations/household-test.json'
 
     with open(class_index_file, 'r') as fin:
         class_mapping = json.loads(fin.read())
@@ -309,9 +309,7 @@ def line_to_map(item, test_mode=False):
         if test_mode:
             return video
         else:
-            template = item['template'].replace('[', '')
-            template = template.replace(']', '')
-            label = int(class_mapping[template])
+            label = int(class_mapping[item['label']])
             return video, label
 
     with open(train_file, 'r') as fin:
@@ -324,7 +322,8 @@ def line_to_map(item, test_mode=False):
 
     with open(test_file, 'r') as fin:
         items = json.loads(fin.read())
-        test_list = [line_to_map(item, test_mode=True) for item in items]
+        #test_list = [line_to_map(item, test_mode=True) for item in items]
+        test_list = [line_to_map(item) for item in items]
 
     splits = ((train_list, val_list, test_list), )
     return splits