Make progress bar + ETA work for stable diffusion

[enolan]

tqdm(enumerate(x)) doesn't let tqdm know how long x is, so the progress bar and ETA don't work. tenumerate conveys the correct information, meaning we get a nice progress bar with an ETA for our diffusions.

[patrickvonplaten]

Hey @enolan,

Would something like this work for you as well: #172 (comment) ?

[anton-l]

@patrickvonplaten to fix this "undefined length" progress bar we'll need to implement #172 with kwargs+total like so:

def progress_bar(self, iterable, **kwargs):
     if self._is_progress_bar_enabled:
         return tqdm(iterable, **kwargs)
     else:
         return iterable

for i, t in tqdm(enumerate(self.scheduler.timesteps), total=self.scheduler.num_inference_steps):

Left a comment on that issue as well

[patrickvonplaten]

Hey @enolan - thanks for the PR, I think we ideally want to follow the official tqdm docs here and simply swap enumerate(...) and tqdm(...) https://pypi.org/project/tqdm/#faq-and-known-issues as is done in #242 .

Hope it's ok to close this PR in favor of #242 :-)

[enolan]

sure, whatever works

…

On Tue, Aug 30, 2022, 02:09 Patrick von Platen ***@***.***> wrote: Closed #236 <#236>. — Reply to this email directly, view it on GitHub <#236 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAPATIUL564PBJ4DBQ7TCDV3XFSZANCNFSM57KJBPOA> . You are receiving this because you were mentioned.Message ID: ***@***.***>