v2.3.0
lib.align: Split lib.align.alignments to smaller modules: - Move update objects to own module - Move Thumbnails to own module - docs update + linting/typing
Cli Restructure + Multi-Mask Select on Extract (#1012)
Preview - Allow mask color/opacity configuration - typing: lib.config - docs: lib.config - bugfix: popup_configure - keep assets on top
Documentation update
bugfix: git communication fixes
Linux - ROCm (AMD) support
Deprecate multi-character cli switches
Update extracted faces to use PNG EXIF data (#1123)
Add Flip Loss Function - Add Flip for AMD and TF - Split Perceptual Loss functions to own modules - Fix allowed input shape for models - Allow GUI tooltip to display at higher width
Faceswap 2.0 (#1045)
Updates: - Remove clipfaker - Add clip to Phaze-A - streamline clip to vision only - update lib.model.networks - Docs and tests update - Phaze-A - add option to flatten bottleneck
bugfix - timelapse image loader multithreading.py - typing + docs
Documentation Update
Core Updates (#1015)
Learning Rate finder (#1341)