TAO toolkit 4.0 actionrecognitionnet training error

When I run this training script from the actionrecognition jupter note book I get a error.
print(“Train RGB only model with PTM”)
!tao action_recognition train
-e $SPECS_DIR/train_rgb_3d_finetune.yaml
-r $RESULTS_DIR/rgb_3d_ptm
-k $KEY
model_config.rgb_pretrained_model_path=$RESULTS_DIR/pretrained/actionrecognitionnet_vtrainable_v1.0/resnet18_3d_rgb_hmdb5_32.tlt
model_config.rgb_pretrained_num_classes=5

Train RGB only model with PTM
usage: tao [-h] {list,stop,info,dataset,deploy,model} …
tao: error: invalid choice: ‘action_recognition’ (choose from ‘list’, ‘stop’, ‘info’, ‘dataset’, ‘deploy’, ‘model’)

Please refer to
https://github.com/NVIDIA/tao_tutorials/blob/main/notebooks/tao_launcher_starter_kit/action_recognition_net/actionrecognitionnet.ipynb.

It is also the same when you download the notebooks via below step.

wget --content-disposition https://api.ngc.nvidia.com/v2/resources/nvidia/tao/tao-getting-started/versions/5.0.0/zip -O getting_started_v5.0.0.zip
unzip -u getting_started_v5.0.0.zip  -d ./getting_started_v5.0.0 && rm -rf getting_started_v5.0.0.zip && cd ./getting_started_v5.0.0

You can also find the user guide in ActionRecognitionNet - NVIDIA Docs

I have already installed the tao toolkit according to the tutorial, as shown in the picture, but it still cannot be trained. Can you help me?

Please download the latest 5.0 notebook.
You can also have a quick check in https://github.com/NVIDIA/tao_tutorials/blob/main/notebooks/tao_launcher_starter_kit/action_recognition_net/actionrecognitionnet.ipynb.

The command line is different.
Should be as below.
! tao model action_recognition train

That’s indeed the problem, thank you!

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.