Pyjcsx

I did not encounter such problem before. Could you successfully run my default training code in VOC dataset?

Hi, sorry for the late reply. You could look in [here](https://github.com/facebookresearch/VMZ/blob/54d93222f80cd23a87c7dbda6044b3fae07b3fda/lib/models/r3d_model.py#L99). When model is r2plus1d, `is_decomposed` is set to `True`. When `is_decomposed` is set `True`, it uses SpatioTemporalConv instead of...

Hi, you may reduce the learning rate.

Did you check your PyTorch version? This whole project is based on PyTorch 0.4.0. Also you could reduce the learning rate.

Hi, thanks for your interest on our work! We are waiting for the company's approval for the code release. Our code/model will be released after the approval.

Hi, thanks for your interest on our work! We are waiting for the company's approval for the code release. Our code/model will be released after the approval.

Hi there, Thanks for your interest in our work! We plan to release our code in the next month. Please stay tuned.