mrq e8f182b634 cleaned up loss calc code (it REALLY hates ignore_loss_for_inputs, but is fine with splitting with loss factors)		2025-02-13 09:35:27 -06:00
..
arch	agony	2025-02-12 00:18:24 -06:00
__init__.py		2024-12-26 21:42:17 -06:00
ar_nar.py	cleanup	2025-02-12 23:36:32 -06:00
base.py	cleaned up loss calc code (it REALLY hates ignore_loss_for_inputs, but is fine with splitting with loss factors)	2025-02-13 09:35:27 -06:00
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