-
Notifications
You must be signed in to change notification settings - Fork 53
simplify screenshot testing #447
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
This seems to work quite well. The ground truths were generated on my RX 570 at home, the largest diff was with the There seems to be no difference between the diffs on Nvidia Tesla V100 and Nvidia A100 on the cluster, at least Nvidia is consistent about one think I guess 😂 EDIT: Just tested on the Nvidia H100 and I get a max diff of 0.02077 on the diffs for all screenshots on a Tesla V100 on the Flatiron cluster for reference:
|
I think that we should figure out how to make the heatmap test screenshot have more of the actual heatmap, not sure why it's small in the top left corner. Or maybe think about getting rid of heatmap since the Texture dimension limit seems to no longer apply with WGPU for the past while, need to investigate this 🤔 |
…plotlib/fastplotlib into simplify-screenshot-testing
use ideas from pygfx/pygfx#684
closes #446
A threshold of 0.2 seemed like it was too lenient, I was able to change the color of one of the circles in the line collection examples and the RMSE was 0.09