You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
You need to provide the model path as resume_path. Since you are providing as a pretrain_path the first convolution kernel is randomly generated regardless of the model size or modality of the inputs.
Hi I have tried to validate the pretrained model with Jester dataset.
Preconditions:
jester_resnext_101_RGB_32.pth
Jester
1.1.0
3.7.3
Test:
python utils/jester_json.py 'annotation_Jester'
to prepare the datasetpython offline_test.py
to start the executionBut the output precision is very poor
Could you please help me to find what am missing to get the proper output?
Regards,
Albin
The text was updated successfully, but these errors were encountered: