Hi there, thanks so much for this implementation! I know this is probably trivial for more advanced users but could you provide some tips or code to accelerate my ability to generate images and movies. Something like what you did for StyleGAN2/ADA would be amazing.

Hello, thanks for your great work.

I tried to train alias-free gan with config-t on the resolution of 1024, by simply editing training.size to 1024 in the config file, but the network seems not converge and the generated result is not good either.

Do I need to modify some other arguments for training on a larger resolution (e.g. 512, 1024)?

Thanks!

Hi, sorry for asking a dumb question. I had a quite hard time (but still failed) to understand the function of filter_parameters in https://github.com/rosinality/alias-free-gan-pytorch/blob/main/model.py#L47

It would be grateful if you can add some comments or explanations to this part. Moreover, especially if we want to change the input dimensions (instead of 16x16 for now), how should we modify this function?

Thanks

Dear Rosinality,

Thank you for you great implementation.

Would you mind providing the pretrained models? Maybe for unaligned FFHQ?

Thank you for your help.

Best Wishes,

Alex

First of all, thanks for taking the time to implement your own version of the alias-free GAN, it's really cool 🙇 I am by no means an expert, but I wanted to propose a couple of changes for the FourierFeature:

• First of all, you were rotating coord_h and then using the rotated value to rotate coord_w, so I fixed that.

• Then, I set the coordinates to the range [-1, 1] so when the rotation is applied it will be performed around the center of the image instead of the edge. Here's an example of a few channels of the Fourier features without and with a rotation applied using a [0, 1] range: And here with a [-1, 1] range:

• Finally, I made it so the translation is dependent on the frequency of every signal since otherwise, the shift in the phase will be different at each waveform. Here are 3 signals and the same signals shifted by 1 unit using a frequency-independent translation value: And here are the same 3 signals when using a frequency-dependent translation: As you can see, by having a frequency-dependent shift we can keep the 3 signals synced.

Hopefully, these changes result in a better relationship between the affine transformation of the FourierFeature and the output images 🤞

Hi! I wanted to try this GAN, but I don't have enough memory to run it. I have a 1080ti 11gb and used training.batch=1 Is there any way to optimize the network so that it fit?

I'm getting the following error in stylegan2/op/fused_act.py at line 66:

RuntimeError: CUDA error: an illegal memory access was encountered

Did not use the docker file and I'm running on torch 1.8.1 and I'm running on a 3090

Thanks for the great contribution.

Slightly off-topic, but do you know how they generate the internal representations in Figure 6 in the original paper? ie. what convention do they use to visualise the channels in the feature maps as RGB?

I tried running this on multiple GPUs and noticed all the processes allocate memory on GPU number 1. This essentially limits me to a batch size of just 1 when each GPU can support closer to a batch size of 4 on their own.

This PR reduces the VRAM usage of the training loop: fixes #16

• Allows 4X increase in batchsize with same number of samples on a 1080TI. Instead of taking up 4X the amount of memory as all the other GPUs, now it only takes up an additional 200mb.
• It moves the inference images to CPU as soon as possible which tells PyTorch immediately frees the underlying GPU memory.
• It also zeros the gradient of the previous backprop so that the memory of those gradients don't have to co-exist with the second forward pass used for inference.
• It ensures the gradients of the generator aren't accidentally stored in the accumulated ema weights (they shouldn't be in the current codebase but this is just an extra sanity check / documentation of desired behavior). Overall, this allows me to fix a batch size of 4 on a 1080TI, where as before I could only fit a batch size of 1. This all works with the default number of samples (16).

@rosinality thank you for always putting out such fantastic work. I have a questions about your training details: How much GPU memory is required, at minimum, to run this implementation of alias free gan? In the paper, the authors mention training on "n NVIDIA DGX-1 with 8 Tesla V100 GPUs", but no mention of how much GPU memory is required. Also, how long did it take you to train?

There is a check that doesn't allow to use conv2d_gradfix with PyTorch 1.9.0 (or later): https://github.com/rosinality/alias-free-gan-pytorch/blob/0e9512888128da58e2bc5945384e2212cab38ff5/stylegan2/op/conv2d_gradfix.py#L85 Is there really some specific functionality that only PyTorch 1.7 and 1.8 have, or is it the check outdated?

README.md says "This implementation contains a lot of my guesses, so I think there are many differences to the official implementations". Is this sentence still accurate?

Traceback (most recent call last): File "train.py", line 404, in conf = load_arg_config(GANConfig) File "/home/user/anaconda3/envs/pytorch/lib/python3.8/site-packages/tensorfn/util/config.py", line 269, in load_arg_config conf = load_config(config_model, args.conf, args.opts, show) File "/home/user/anaconda3/envs/pytorch/lib/python3.8/site-packages/tensorfn/util/config.py", line 257, in load_config conf = config_model(**read_config(config, overrides=overrides)) File "/home/user/anaconda3/envs/pytorch/lib/python3.8/site-packages/tensorfn/util/config.py", line 25, in read_config json_str = _jsonnet.evaluate_file(config_file) AttributeError: 'NoneType' object has no attribute 'evaluate_file'

data format.

-Data
--Images
---1.jpg
|
|
|
---N.jpg

After reflecting from the official pytorch code, this project doesn't work in config-r.jsonnet setting. For detail, an error occurs when kernel_size is 1. I found a bug in padding size, so I modify by reflecting the official code.

I´ve been trying to run the repo via colab. Everything goes fine until running the train script, then the error below is thrown. Has the repo been recently updated or am I missing some dependency or some other detail to be considered? Any clue how to solve this?

IndexError Traceback (most recent call last) /usr/local/lib/python3.7/dist-packages/pyparsing.py in _parseNoCache(self, instring, loc, doActions, callPreParse) 1682 try: -> 1683 loc, tokens = self.parseImpl(instring, preloc, doActions) 1684 except IndexError: 8 frames IndexError: string index out of range During handling of the above exception, another exception occurred: ParseException Traceback (most recent call last) ParseException: Expected {: ... | {{{"=" | ":" | "+="} - [Suppress:({{{"#" | "//"} - SkipTo:({Suppress:(W:( )) | StringEnd})} | Suppress:(W:( ))})]... -} ConcatenatedValueParser:([{{{{Suppress:({[Suppress:(W:( ,))] {"#" | "//"} - SkipTo:({Suppress:(W:( )) | StringEnd})}) | {Suppress:("include") {Re:('"(?:[^"\\\n]|\\.)"[ \t]') | {{"url" | "file" | "package"} - Suppress:("(") - Re:('"(?:[^"\\\n]|\\.)"[ \t]') - Suppress:(")")} | {"required" - Suppress:("(") - {Re:('"(?:[^"\\\n]|\\.)"[ \t]') | {{"url" | "file" | "package"} - Suppress:("(") - Re:('"(?:[^"\\\n]|\\.)"[ \t]') - Suppress:(")")}} - Suppress:(")")}}} | Re:('[ \t]\$\{[^\}]+\}[ \t]') | : ...} | Forward: {{{{Suppress:("[") -} ListParser:({{ConcatenatedValueParser:([{{{{Suppress:({[Suppress:(W:( ,))] {"#" | "//"} - SkipTo:({Suppress:(W:( )) | StringEnd})}) | {Suppress:("include") {Re:('"(?:[^"\\\n]|\\.)"[ \t]') | {{"url" | "file" | "package"} - Suppress:("(") - Re:('"(?:[^"\\\n]|\\.)"[ \t]') - Suppress:(")")} | {"required" - Suppress:("(") - {Re:('"(?:[^"\\\n]|\\.)"[ \t]') | {{"url" | "file" | "package"} - Suppress:("(") - Re:('"(?:[^"\\\n]|\\.)"[ \t]') - Suppress:(")")}} - Suppress:(")")}}} | Re:('[ \t]\$\{[^\}]+\}[ \t]') | : ...} | : ...} | {{{{{{{{W:(0123...) Suppress:([]...)} {"ns" ^ "nano" ^ "nanos" ^ "nanosecond" ^ "nanoseconds" ^ "us" ^ "micro" ^ "micros" ^ "microsecond" ^ "microseconds" ^ "ms" ^ "milli" ^ "millis" ^ "millisecond" ^ "milliseconds" ^ "s" ^ "second" ^ "seconds" ^ "m" ^ "minute" ^ "minutes" ^ "h" ^ "hour" ^ "hours" ^ "w" ^ "week" ^ "weeks" ^ "d" ^ "day" ^ "days" ^ "mo" ^ "month" ^ "months" ^ "y" ^ "year" ^ "years"}} Supp} | {{{{{{{{W:(0123...) Suppress:([]...)} {"ns" ^ "nano" ^ "nanos" ^ "nanosecond" ^ "nanoseconds" ^ "us" ^ "micro" ^ "micros" ^ "microsecond" ^ "microseconds" ^ "ms" ^ "milli" ^ "millis" ^ "millisecond" ^ "milliseconds" ^ "s" ^ "second" ^ "seconds" ^ "m" ^ "minute" ^ "minutes" ^ "h" ^ "hour" ^ "hours" ^ "w" ^ "week" ^ "weeks" ^ "d" ^ "day" ^ "days" ^ "mo" ^ "month" ^ "months" ^ "y" ^ "year" ^ "years"}} Suppress:(WordEnd)} | Re:('[+-]?(\d*\.\d+|\d+(\.\d+)?)([eE][+\-]?\d+)?(?=$|[ \t]([\$\}\],#\n\r]|//))')} | "true"} | "false"} | "null"} | {{Re:('""".?""""') | Re:('"(?:[^"\\\n]|\\.)"[ \t]*')} | Re:('(... ))})}]...)}}, found end of text (at char 43), (line:1, col:44)