Muhammad Rizqi Nur
Muhammad Rizqi Nur
This is a stripped down PR of #4214, cherry picked and added a few changes There was still some issue or room for improvements in #3986: 1. restore_base_vae was actually...
Here I'm using canny edge. No lora, hypernet, or embedding used.  Here's an example with batch count=4, weight=1, and guidance=0.94 (or higher)     But when I...
Windows Defender flagged zluda.exe (Version 3) as a Trojan:Win32/Wacatac.B!ml  [3 flags on VirusTotal](https://www.virustotal.com/gui/file/8c06c8a80000bdf1f3ce2b58c5a577a7b84393505b5aedbdbd8cc940b64820dc) It should be a false positive, right?
### Motivation Simultaneously optimizing a general hyperparameter set, such as activation functions, and dataset-specific hyperparameter set, such as hidden dimension size. Doing the search in parallel for different datasets. ###...
**Is your feature request related to a problem? Please describe.** Torchinfo summary is very useful to debug shapes and module chain. However, it doesn't work when it encounters error. Often,...
Fix #143
Fixes #143 In case cat_name isn't provided, cat_id will be converted to string and used as cat_name. If this gets merged, this behavior should probably be documented though. Next, specifying...
Cat_names is optional, but ShowClassSplits() have to use cat_names. This makes ShowClassSplits return empty dataframe  It should just use cat_id if cat_names is not provided. I'll try fixing
# Feature Request I have a lot of parameters and only want to view some of them in parallel coordinates. However they're all enabled by default. Unchecking almost every parameters...
I'd like to use texture with automatic UV mapping like in minecraft grass
Issue #6553, resume=True doesn't actually enforce caching. It was caused by resuming a model that was trained with caching. This PR allows overriding cache and epochs. It also removes and...