mrq 09e9438941 ugh		2025-03-25 23:24:01 -05:00
..
arch	fixed errant index error (although it makes me wonder if my segmented masking is still flawed)	2025-03-21 23:41:34 -05:00
__init__.py	nothing could go wrong part 2 (reverted and rewrote commits since there was a nasty regression)	2025-03-25 23:06:16 -05:00
ar_nar_v2.py	nothing could go wrong part 2 (reverted and rewrote commits since there was a nasty regression)	2025-03-25 23:06:16 -05:00
ar_nar.py	nothing could go wrong part 2 (reverted and rewrote commits since there was a nasty regression)	2025-03-25 23:06:16 -05:00
base_v2.py	ugh	2025-03-25 23:24:01 -05:00
base.py	nothing could go wrong part 2 (reverted and rewrote commits since there was a nasty regression)	2025-03-25 23:06:16 -05:00
lora.py	naive model offloading support (handles automatically splitting parts of the model to requested device per memory constraints, either inferred or requested in the yaml, input tensors are automatically migrated to the right device, it SEEMS to work for training under the test trainer when split between GPU and CPU) (this was specifically only because that Flux imagegen model released so I can test it there)	2024-08-01 20:12:06 -05:00