Li Bo

I think my above question is related to the **cycle consistency loss**. Can we just load a pre-trained model **fs** to compute task loss between images from the _source_ and...

> @Luodian Hey, have you solved this problem or got any ideas? I am just starting to try to implement on my own:) Of course you could implement on your...

@jhoffman I encountered the same problem. the domacc value is set 60. But until the end of training, the discriminator network's accuracy can't reach that high.

I tried to set the value at 55, it works better but I am not sure whether it's best. Another question, during training, the mIOU keeps degrading from 90 to...

Thank you~

@wang338 Usually in GTAV you will get a suitable `D` model after ~4000 iterations. It will jump out of training after 2950 iterations if you are using `cityscapes` as the...

Hoping I clarified it clearly~If you have other questions probably we can go through QQ or Wechat.

thanks! I think it's worthy doing.

Hi I think this happens because the default `gate_noise` value in `load_importance_loss` is `0.0`. And if we do ```python normal = Normal(0, 0.0) ``` it's weird, why we have a...

If I preset gate_noise to 1.0, I think the code run without problems but I am not sure if it's numerically correct? ```python gate_type={'type': 'top', 'k': 2, 'fp32_gate': False, 'gate_noise':...