Same outputs after model.eval()

[farsal]

Hi, I have some issues when using model on validation data, model.eval() gives the same output for any validation data, Can you please help me with this?

[WinstonHuTiger]

@farsal did you solve this, I face the same problem.

[nooppi18]

I am facing the same problem.

[pranshu97]

Did anyone find a solution for this issue?
@shijianjian @farsal @WinstonHuTiger @nooppi18

[shijianjian]

Sorry that I did not aware of those issues in this repo. I currently do not actively maintain this repo. So, if you have resolved this, drop me a PR and at me. I will merge.

I do not check much on the code. Presumable, if model.train() worked but not model.eval(), I suspect something went wrong with train/test conditional layers as BatchNorm, dropout, etc. Good luck!

[AmokraneMancer]

same issue, did any one find a solution ?

[myxik]

please disable model.eval() and use torch.no_grad()