search

JusperLee / CTCNet

An Audio-Visual Speech Separation Model Inspired by Cortico-Thalamo-Cortical Circuits
Apache License 2.0
69 stars 16 forks source link

A probable typo in the code? #12

Open DRfn1234 opened 8 months ago

DRfn1234 commented 8 months ago

Hello!When I try to train the model, there is an error said AttributeError: 'VideoBlock' object has no attribute 'get_block_block' and there is no get_block_blockin class VideoBlock in videosubnetwork.py. Can you help fix this? Thanks in advance.

DRfn1234 commented 8 months ago

I change the get_block_block to get_video_block and hope it'll be correct. But another error occurs that TypeError: stack(): argument 'tensors' (position 1) must be tuple of Tensors, not NoneType in def on_validation_epoch_end(self): after I add validation_step_outputs=None indef initin order to solve the problem in #8. Do you know how to fix it?

JusperLee commented 8 months ago

My guess would be that the input is none causing this error. You can test the model individually to see if the inputs and outputs are correct.

DRfn1234 commented 8 months ago

Which value should I set to the validation_step_outputs? Because it requires an input.