25 Commits

Author	SHA1	Message	Date
Kohya S	81c0c965a2	faster block swap	2024-11-05 21:22:42 +09:00
Kohya S	623017f716	refactor SD3 CLIP to transformers etc.	2024-10-24 19:49:28 +09:00
Kohya S	2500f5a798	fix latents caching not working closes #1696	2024-10-15 07:16:34 +09:00
kohya-ss	5bb9f7fb1a	Merge branch 'sd3' into multi-gpu-caching	2024-10-13 11:52:42 +09:00
Kohya S	e277b5789e	Update FLUX.1 support for compact models	2024-10-12 21:49:07 +09:00
kohya-ss	c80c304779	Refactor caching in train scripts	2024-10-12 20:18:41 +09:00
Kohya S	83e3048cb0	load Diffusers format, check schnell/dev	2024-10-06 21:32:21 +09:00
Kohya S	ba08a89894	call optimizer eval/train for sample_at_first, also set train after resuming closes #1667	2024-10-04 20:35:16 +09:00
Kohya S	da94fd934e	fix typos	2024-09-26 08:27:48 +09:00
Kohya S	56a7bc171d	new block swap for FLUX.1 fine tuning	2024-09-26 08:26:31 +09:00
Kohya S	1286e00bb0	fix to call train/eval in schedulefree #1605	2024-09-18 21:31:54 +09:00
Kohya S	a2ad7e5644	blocks_to_swap=0 means no swap	2024-09-17 21:42:14 +09:00
Plat	a823fd9fb8	Improve wandb logging (#1576) ... * fix: wrong training steps were recorded to wandb, and no log was sent when logging_dir was not specified * fix: checking of whether wandb is enabled * feat: log images to wandb with their positive prompt as captions * feat: logging sample images' caption for sd3 and flux * fix: import wandb before use	2024-09-11 22:21:16 +09:00
Kohya S	2889108d85	feat: Add --cpu_offload_checkpointing option to LoRA training	2024-09-05 20:58:33 +09:00
Kohya S	d9129522a6	set dtype before calling ae closes #1562	2024-09-05 12:20:07 +09:00
Kohya S	8ecf0fc4bf	Refactor code to ensure args.guidance_scale is always a float #1525	2024-08-29 22:10:57 +09:00
Kohya S	a4d27a232b	Fix --debug_dataset to work.	2024-08-22 19:55:31 +09:00
kohya-ss	98c91a7625	Fix bug in FLUX multi GPU training	2024-08-22 12:37:41 +09:00
Kohya S	e1cd19c0c0	add stochastic rounding, fix single block	2024-08-21 21:04:10 +09:00
Kohya S	7e459c00b2	Update T5 attention mask handling in FLUX	2024-08-21 08:02:33 +09:00
Kohya S	486fe8f70a	feat: reduce memory usage and add memory efficient option for model saving	2024-08-19 22:30:24 +09:00
Kohya S	6e72a799c8	reduce peak VRAM usage by excluding some blocks to cuda	2024-08-19 21:55:28 +09:00
Kohya S	ef535ec6bb	add memory efficient training for FLUX.1	2024-08-18 16:54:18 +09:00
kohya-ss	25f77f6ef0	fix flux fine tuning to work	2024-08-17 15:54:32 +09:00
Kohya S	400955d3ea	add fine tuning FLUX.1 (WIP)	2024-08-17 15:36:18 +09:00