zzmartin

I think it is the taming-transformers repo too large; I download the zip of it, unzip and run the setup.py to install. Hope works for you.

And need to run: `pip install -e .` in the taming-transformers folder.

I got same issue. I used isv and got mean:16.668, std: 0.4938

@anonymous202203 Hi, have you figured it out?

@anonymous202203 do you have its pretrained model on ucf101? Their implementation is a little bit confused if without pretrained model config

@anonymous202203 can you share your parameters of tgan2? I test it on the original UCF-101, the IS only got around 30, which is supposed to be 60.

@universome yeah, you are right. when setting to 13320, it got almost 60, thx.

@universome Well, from LDVD GAN, if set resolution to 128, the IS should be around 80-90. Why u got 84 when set resolution to 256? Shouldn't it be bigger? 

@universome you are right, there should be no difference when resolution higher than 112. I wonder how you process the data? I chose central 32 frame of the each video(due...

I first crop UCF to 240*240, and store as MP4. Then I use your script to convert UCF videos to JPEG. I tried store 128 and 256 images, both only...