search

microsoft / View-Adaptive-Neural-Networks-for-Skeleton-based-Human-Action-Recognition

View Adaptive Neural Networks for High Performance Skeleton-based Human Action Recognition
MIT License
127 stars 58 forks source link

Mismatch between data loader returning items and items required by the model #3

Closed sksenthilkumar closed 4 years ago

sksenthilkumar commented 4 years ago

For the given code to work, we need a large memory space, because the dataset is loaded completely at once. Hence I'm rewriting the code for my requirement. During which I noticed that:

The VA-CNN model requires two arguments input joints and maxmin value (line).

at the same time, the data loader created during the training initialization does not return maxmin value (this line & this line)

please help me by clarifying the mismatch. Thanks

sksenthilkumar commented 4 years ago

I got it. It is given as a part of the collate function in data loader (here).