Polish modelzoo readme and update `getting_started.md` url

04f44fa7 · linjintao · lizz · d2bce4ba · 04f44fa7 · 04f44fa7
10 changed file
--- a/configs/localization/bmn/README.md
+++ b/configs/localization/bmn/README.md
 # BMN

+## Model Zoo
+
+### ActivityNet feature
+
 |config | pretrain | AR@100| AUC | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[bmn_400x100_9e_2x8_activitynet_feature](bmn_400x100_2x8_9e_activitynet_feature.py) | None |75.28|67.22|5420|3.27|[ckpt]()| [log]()|

-## Preparation
-### Data
+## Data
 1. Put the rescaled feature data folder `csv_mean_100` under `$MMACTION/data/activitynet_feature_cuhk/`.

    The raw feature data could be found at [here](https://github.com/wzmsltw/BSN-boundary-sensitive-network).
@@ -31,7 +34,7 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 Put the `tem_best.pth.tar` and `pem_best.pth.tar` under `checkpoints/`.
 The ckpts could be found at [here]().


--- a/configs/localization/bsn/README.md
+++ b/configs/localization/bsn/README.md
 # BSN

+## Model Zoo
+
+### ActivityNet feature
+
 |config | pretrain | AR@100| AUC | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |bsn_400x100_1x16_20e_activitynet_feature | None |74.65|66.45|41(TEM)+25(PEM)|0.074(TEM)+0.036(PEM)|[ckpt_tem]() [ckpt_pem]| [log_tem]() [log_pem]()|

-## Preparation
-### Data
+## Data
 1. Put the rescaled feature data folder `csv_mean_100` under `$MMACTION/data/activitynet_feature_cuhk/`.

    The raw feature data could be found at [here](https://github.com/wzmsltw/BSN-boundary-sensitive-network).
@@ -31,7 +34,7 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 1. Put the `tem_best.pth.tar` and `pem_best.pth.tar` under `checkpoints/`.

    The ckpts could be found at [here]() (TODO).
@@ -55,7 +58,7 @@ Examples:
    python tools/train.py configs/localization/bsn/bsn_pem_400x100_1x16_20e_activitynet_feature.py
    ```

-For more details and optional arguments infos, you can refer to **Training setting** part in [GETTING_START](../../../docs/getting_started.md).
+For more details and optional arguments infos, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

 ## Inference
 You can use the following commands to inference a model.
@@ -129,4 +132,4 @@ Examples:
    ```shell
    python tools/test.py configs/localization/bsn/bsn_400x100_1x16_20e_activitynet_feature.py checkpoints/SOME_CHECKPOINT.pth --eval AR@AN --out results.json
    ```
-For more details and optional arguments infos, you can refer to **Test a dataset** part in [GETTING_START](../../../docs/getting_started.md).
+For more details and optional arguments infos, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/configs/recognition/i3d/README.md
+++ b/configs/recognition/i3d/README.md
 # I3D

-## Kinetics400
+## Model Zoo
+
+### Kinetics-400

 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[i3d_r34_32x2x1_100e_kinetics400_rgb](i3d_r34_32x2x1_100e_kinetics400_rgb.py) | ResNet34 |68.37|88.15| x | x | [ckpt]() | [log]()|
 |[i3d_r50_32x2x1_100e_kinetics400_rgb](i3d_r50_32x2x1_100e_kinetics400_rgb.py) | ResNet50 |72.68|90.78| x | x | [ckpt]() | [log]()|
 |[i3d_r50_dense_32x2x1_100e_kinetics400_rgb](i3d_r50_dense_32x2x1_100e_kinetics400_rgb.py) | ResNet50 |72.77|90.57| x | x | [ckpt]() | [log]()|
 |[i3d_r50_fast_32x2x1_100e_kinetics400_rgb](i3d_r50_fast_32x2x1_100e_kinetics400_rgb.py) | ResNet50 |72.32|90.72| x | x | [ckpt]() | [log]()|
 |[i3d_r50_video_3d_32x2x1_100e_kinetics400_rgb](i3d_r50_video_32x2x1_100e_kinetics400_rgb.py) | ResNet50 | x | x | x | x | [ckpt]() | [log]()|

-### Data
+## Data

 1. Make a dataset folder under the path `$MMACTION/data`.
 2. Put the data sub folders (commonly including `rawframes_train/` + `rawframes_val/` or `video_train` + `video_val`) under `$MMACTION/data/dataset_name`.
@@ -42,35 +44,35 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 Put the checkpoint required under `$MMACTION/checkpoints`. The checkpoints can be found at [here]().

-### Train
+## Train
 You can use the following command to train a model.
 ```shell
 python tools/train.py ${CONFIG_FILE} [optional arguments]
 ```

-Example: train I3D model on Kinetics400 dataset in a deterministic option with periodic validation.
+Example: train I3D model on Kinetics-400 dataset in a deterministic option with periodic validation.
 ```shell
 python tools/train.py configs/recognition/i3d/i3d_r50_32x2x1_100e_kinetics400_rgb.py \
    --work_dir work_dirs/i3d_r50_32x2x1_100e_kinetics400_rgb \
    --validate --seed 0 --deterministic
 ```

-For more details, you can refer to **Training setting** part in [GETTING_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

-### Test
+## Test
 You can use the following command to test a model.
 ```shell
 python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [optional arguments]
 ```

-Example: test I3D model on Kinetics400 dataset and dump the result to a json file.
+Example: test I3D model on Kinetics-400 dataset and dump the result to a json file.
 ```shell
 python tools/test.py configs/recognition/i3d/i3d_r50_32x2x1_100e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/configs/recognition/r2plus1d/README.md
+++ b/configs/recognition/r2plus1d/README.md
 # R2plus1D

-## Kinetics400
+## Model Zoo
+
+### Kinetics-400

 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[r2plus1d_r34_8x8x1_180e_kinetics400_rgb](r2plus1d_r34_8x8x1_180e_kinetics400_rgb.py) | None |68.68|88.36|5014|0.68|[ckpt]()| [log]()|
 |[r2plus1d_r34_32x2x1_180e_kinetics400_rgb](r2plus1d_r34_32x2x1_180e_kinetics400_rgb.py) | None |74.60|91.59|9003|1.23| [ckpt]() | [log]()|
 |[r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb](r2plus1d_r34_video_8x8x1_180e_kinetics400_rgb.py) | None |x|x|x|x| [ckpt]() | [log]()|

-### Data
+## Data

 1. Make a dataset folder under the path `$MMACTION/data`.
 2. Put the data sub folders (commonly including `rawframes_train/` + `rawframes_val/` or `video_train` + `video_val`) under `$MMACTION/data/dataset_name`.
@@ -40,35 +42,35 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 Put the checkpoint required under `$MMACTION/checkpoints`. The checkpoints can be found at [here]().

-### Train
+## Train
 You can use the following command to train a model.
 ```shell
 python tools/train.py ${CONFIG_FILE} [optional arguments]
 ```

-Example: train R(2+1)D model on Kinetics400 dataset in a deterministic option with periodic validation.
+Example: train R(2+1)D model on Kinetics-400 dataset in a deterministic option with periodic validation.
 ```shell
 python tools/train.py configs/recognition/r2plus1d/r2plus1d_r34_3d_8x8x1_180e_kinetics400_rgb.py \
    --work_dir work_dirs/r2plus1d_r34_3d_8x8x1_180e_kinetics400_rgb \
    --validate --seed 0 --deterministic
 ```

-For more details, you can refer to **Training setting** part in [GETTING_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

-### Test
+## Test
 You can use the following command to test a model.
 ```shell
 python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [optional arguments]
 ```

-Example: test R(2+1)D model on Kinetics400 dataset and dump the result to a json file.
+Example: test R(2+1)D model on Kinetics-400 dataset and dump the result to a json file.
 ```shell
 python tools/test.py configs/recognition/r2plus1d/r2plus1d_r34_3d_8x8x1_180e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/configs/recognition/slowfast/README.md
+++ b/configs/recognition/slowfast/README.md
 # SlowFast

-## Kinetics400
+## Model Zoo
+
+### Kinetics-400

 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[slowfast_r50_4x16x1_256e_kinetics400_rgb](slowfast_r50_4x16x1_256e_kinetics400_rgb.py) | None |75.3|92.2|6250|0.826|[ckpt]()| [log]()|
 |[slowfast_r50_8x8x1_256e_kinetics400_rgb](slowfast_r50_8x8x1_256e_kinetics400_rgb.py) | None |76.36|92.56|9159|1.032| [ckpt]() | [log]()|
 |[slowfast_r50_video_4x16x1_256e_kinetics400_rgb](slowfast_r50_video_4x16x1_256e_kinetics400_rgb.py) | None |x|x|x|x| [ckpt]() | [log]()|

-### Data
+## Data

 1. Make a dataset folder under the path `$MMACTION/data`.
 2. Put the data sub folders (commonly including `rawframes_train/` + `rawframes_val/` or `video_train` + `video_val`) under `$MMACTION/data/dataset_name`.
@@ -40,35 +42,35 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 Put the checkpoint required under `$MMACTION/checkpoints`. The checkpoints can be found at [here]().

-### Train
+## Train
 You can use the following command to train a model.
 ```shell
 python tools/train.py ${CONFIG_FILE} [optional arguments]
 ```

-Example: train SlowFast model on Kinetics400 dataset in a deterministic option with periodic validation.
+Example: train SlowFast model on Kinetics-400 dataset in a deterministic option with periodic validation.
 ```shell
 python tools/train.py configs/recognition/slowfast/slowfast_r50_3d_4x16x1_256e_kinetics400_rgb.py \
    --work_dir work_dirs/slowfast_r50_3d_4x16x1_256e_kinetics400_rgb \
    --validate --seed 0 --deterministic
 ```

-For more details, you can refer to **Training setting** part in [GETTING_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

-### Test
+## Test
 You can use the following command to test a model.
 ```shell
 python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [optional arguments]
 ```

-Example: test SlowFast model on Kinetics400 dataset and dump the result to a json file.
+Example: test SlowFast model on Kinetics-400 dataset and dump the result to a json file.
 ```shell
 python tools/test.py configs/recognition/slowfast/slowfast_r50_3d_4x16x1_256e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/configs/recognition/tin/README.md
+++ b/configs/recognition/tin/README.md
 # TIN

-## Kinetics400
+## Model Zoo
+
+### Kinetics-400
 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[tin_r50_1x1x8_35e_kinetics400_rgb](tin_r50_1x1x8_35e_kinetics400_rgb.py) | ResNet50 |69.44|89.19| x | x | [ckpt]() | [log]()|
 |[tin_r50_finetune_1x1x8_35e_kinetics400_rgb](tin_r50_finetune_1x1x8_35e_kinetics400_rgb.py) | TSM |71.00|89.98| x | x | [ckpt]() | [log]()|
 |[tin_r50_video_2d_1x1x8_35e_kinetics400_rgb](tin_r50_video_1x1x8_35e_kinetics400_rgb.py) | ResNet50 | x | x | x | x | [ckpt]() | [log]()|

-## Something-Something V1&V2
+### Something-Something V1
 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[tin_r50_1x1x8_35e_sthv1_rgb](tin_r50_1x1x8_35e_sthv1_rgb.py) | ResNet50 |41.59|71.94| x | x | [ckpt]() | [log]()|
-|[tin_r50_1x1x8_35e_sthv2_rgb](tin_r50_1x1x8_35e_sthv2_rgb.py) | ResNet50 |53.08|82.02| x | x | [ckpt]() | [log]()|

+### Something-Something V2
+|config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
+|-|-|-|-|-|-|-|-|
+|[tin_r50_1x1x8_35e_sthv2_rgb](tin_r50_1x1x8_35e_sthv2_rgb.py) | ResNet50 |53.08|82.02| x | x | [ckpt]() | [log]()|

-### Data
+## Data

 1. Make a dataset folder under the path `$MMACTION/data`.
 2. Put the data sub folders (commonly including `rawframes_train/` + `rawframes_val/` or `video_train` + `video_val`) under `$MMACTION/data/dataset_name`.
@@ -46,35 +51,35 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 Put the checkpoint required under `$MMACTION/checkpoints`. The checkpoints can be found at [here]().

-### Train
+## Train
 You can use the following command to train a model.
 ```shell
 python tools/train.py ${CONFIG_FILE} [optional arguments]
 ```

-Example: train TIN model on Kinetics400 dataset in a deterministic option with periodic validation.
+Example: train TIN model on Kinetics-400 dataset in a deterministic option with periodic validation.
 ```shell
 python tools/train.py configs/recognition/tin/tin_r50_1x1x8_35e_kinetics400_rgb.py \
    --work_dir work_dirs/tin_r50_1x1x8_35e_kinetics400_rgb \
    --validate --seed 0 --deterministic
 ```

-For more details, you can refer to **Training setting** part in [GETTING_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

-### Test
+## Test
 You can use the following command to test a model.
 ```shell
 python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [optional arguments]
 ```

-Example: test TIN model on Kinetics400 dataset and dump the result to a json file.
+Example: test TIN model on Kinetics-400 dataset and dump the result to a json file.
 ```shell
 python tools/test.py configs/recognition/tin/tin_r50_1x1x8_35e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/configs/recognition/tsm/README.md
+++ b/configs/recognition/tsm/README.md
 # TSM

-## Kinetics400
+## Model Zoo
+
+### Kinetics-400
 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[tsm_r50_1x1x8_100e_kinetics400_rgb](tsm_r50_1x1x8_50e_kinetics400_rgb.py) | ResNet50 |70.24|89.56| x | x | [ckpt]() | [log]()|
 |[tsm_r50_dense_1x1x8_100e_kinetics400_rgb](tsm_r50_dense_1x1x8_50e_kinetics400_rgb.py) | ResNet50 |71.84|90.18| x | x | [ckpt]() | [log]()|
 |[tsm_r50_1x1x16_50e_kinetics400_rgb](tsm_r50_1x1x16_50e_kinetics400_rgb.py) | ResNet50 |71.69|90.4| x | x | [ckpt]() | [log]()|
 |[tsm_r50_video_1x1x8_100e_kinetics400_rgb](tsm_r50_video_1x1x8_100e_kinetics400_rgb.py) | ResNet50 | x | x | x | x | [ckpt]() | [log]()|

-## Something-Something V1&V2
+### Something-Something V1
 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
-|-|-|-|-|-|-|- | -|
+|-|-|-|-|-|-|-|-|
 |[tsm_r50_1x1x8_50e_sthv1_rgb](tsm_r50_1x1x8_50e_sthv1_rgb.py) | ResNet50 |44.62|75.51| x | x | [ckpt]() | [log]()|
 |[tsm_r50_1x1x16_50e_sthv1_rgb](tsm_r50_1x1x16_50e_sthv1_rgb.py) | ResNet50 |43.81|74.73| x | x | [ckpt]() | [log]()|
 |[tsm_r101_1x1x8_50e_sthv1_rgb](tsm_r101_1x1x8_50e_sthv1_rgb.py) | ResNet50 |46.41|74.07| x | x | [ckpt]() | [log]()|
+
+### Something-Something V2
+|config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
+|-|-|-|-|-|-|-|-|
 |[tsm_r50_1x1x8_50e_sthv2_rgb](tsm_r50_1x1x8_50e_sthv2_rgb.py) | ResNet50 |59.91|84.61| x | x | [ckpt]() | [log]()|
 |[tsm_r50_1x1x16_50e_sthv2_rgb](tsm_r50_1x1x16_50e_sthv2_rgb.py) | ResNet50 |56.10|84.43| x | x | [ckpt]() | [log]()|
 |[tsm_r101_1x1x8_50e_sthv2_rgb](tsm_r101_1x1x8_50e_sthv2_rgb.py) | ResNet50 | x | x | x | x | [ckpt]() | [log]()|

-### Data
+## Data

 1. Make a dataset folder under the path `$MMACTION/data`.
 2. Put the data sub folders (commonly including `rawframes_train/` + `rawframes_val/` or `video_train` + `video_val`) under `$MMACTION/data/dataset_name`.
@@ -50,35 +56,35 @@ mmaction
 ...
 ```

-### Checkpoint
+## Checkpoint
 Put the checkpoint required under `$MMACTION/checkpoints`. The checkpoints can be found at [here]().

-### Train
+## 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 Kinetics400 dataset in a deterministic option with periodic validation.
+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_100e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

-### Test
+## 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 Kinetics400 dataset and dump the result to a json file.
+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_100e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/configs/recognition/tsn/README.md
+++ b/configs/recognition/tsn/README.md
@@ -8,7 +8,7 @@
 |-|-|-|-|-|-|-|-|
 |[tsn_r50_1x1x3_100e_ucf101_rgb](/configs/recognition/tsn/tsn_r50_1x1x3_100e_ucf101_rgb.py) | ResNet50 |80.12|96.09| x | x | [ckpt]() | [log]()|

-### Kinetics
+### Kinetics-400

 |config | pretrain | top1 acc| top5 acc | gpu_mem(M) | iter time(s) | ckpt | log|
 |-|-|-|-|-|-|-|-|
@@ -88,14 +88,14 @@ You can use the following command to train a model.
 python tools/train.py ${CONFIG_FILE} [optional arguments]
 ```

-Example: train TSN model on Kinetics400 dataset in a deterministic option with periodic validation.
+Example: train TSN model on Kinetics-400 dataset in a deterministic option with periodic validation.
 ```shell
 python tools/train.py configs/recognition/tsn/tsn_r50_1x1x3_100e_kinetics400_rgb.py \
    --work_dir work_dirs/tsn_r50_1x1x3_100e_kinetics400_rgb \
    --validate --seed 0 --deterministic
 ```

-For more details, you can refer to **Training setting** part in [GETTING_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Training setting** part in [getting_started](../../../docs/getting_started.md).

 ## Test

@@ -104,11 +104,11 @@ You can use the following command to test a model.
 python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [optional arguments]
 ```

-Example: test TSN model on Kinetics400 dataset and dump the result to a json file.
+Example: test TSN model on Kinetics-400 dataset and dump the result to a json file.
 ```shell
 python tools/test.py configs/recognition/tsn/tsn_r50_1x1x3_100e_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_START](../../../docs/GETTING_STARTED.md).
+For more details, you can refer to **Test a dataset** part in [getting_started](../../../docs/getting_started.md).
--- a/docs/getting_started.md
+++ b/docs/getting_started.md
@@ -31,7 +31,7 @@ For using custom datasets, please refer to [Tutorial 2: Adding New Dataset](tuto

 ## Inference with Pre-Trained Models

-We provide testing scripts to evaluate a whole dataset (Kinetics400, Something-Something V1&V2, (Multi-)Moments in Time, etc.),
+We provide testing scripts to evaluate a whole dataset (Kinetics-400, Something-Something V1&V2, (Multi-)Moments in Time, etc.),
 and provide some high-level apis for easier integration to other projects.

 ### Test a dataset

--- a/docs/tutorials/finetune.md
+++ b/docs/tutorials/finetune.md
@@ -17,7 +17,7 @@ The `num_classes` in the `cls_head` need to be changed to the class number of th
 The weights of the pre-trained models are reused except for the final prediction layer.
 So it is safe to change the class number.
 In our case, UCF101 has 101 classes.
-So we change it from 400 (class number of Kinetics400) to 101.
+So we change it from 400 (class number of Kinetics-400) to 101.

 ```python
 model = dict(