| The cache for model files in Transformers v4.22.0 has been updated. Migrating your old cache. This is a one-time only operation. You can interrupt this and resume the migration later on by calling `transformers.utils.move_cache()`. | |
| Running 1 job | |
| 0it [00:00, ?it/s] 0it [00:00, ?it/s] | |
| /usr/local/lib/python3.10/dist-packages/controlnet_aux/mediapipe_face/mediapipe_face_common.py:7: UserWarning: The module 'mediapipe' is not installed. The package will have limited functionality. Please install it using the command: pip install 'mediapipe' | |
| warnings.warn( | |
| /usr/local/lib/python3.10/dist-packages/controlnet_aux/segment_anything/modeling/tiny_vit_sam.py:654: UserWarning: Overwriting tiny_vit_5m_224 in registry with controlnet_aux.segment_anything.modeling.tiny_vit_sam.tiny_vit_5m_224. This is because the name being registered conflicts with an existing name. Please check if this is not expected. | |
| return register_model(fn_wrapper) | |
| /usr/local/lib/python3.10/dist-packages/controlnet_aux/segment_anything/modeling/tiny_vit_sam.py:654: UserWarning: Overwriting tiny_vit_11m_224 in registry with controlnet_aux.segment_anything.modeling.tiny_vit_sam.tiny_vit_11m_224. This is because the name being registered conflicts with an existing name. Please check if this is not expected. | |
| return register_model(fn_wrapper) | |
| /usr/local/lib/python3.10/dist-packages/controlnet_aux/segment_anything/modeling/tiny_vit_sam.py:654: UserWarning: Overwriting tiny_vit_21m_224 in registry with controlnet_aux.segment_anything.modeling.tiny_vit_sam.tiny_vit_21m_224. This is because the name being registered conflicts with an existing name. Please check if this is not expected. | |
| return register_model(fn_wrapper) | |
| /usr/local/lib/python3.10/dist-packages/controlnet_aux/segment_anything/modeling/tiny_vit_sam.py:654: UserWarning: Overwriting tiny_vit_21m_384 in registry with controlnet_aux.segment_anything.modeling.tiny_vit_sam.tiny_vit_21m_384. This is because the name being registered conflicts with an existing name. Please check if this is not expected. | |
| return register_model(fn_wrapper) | |
| /usr/local/lib/python3.10/dist-packages/controlnet_aux/segment_anything/modeling/tiny_vit_sam.py:654: UserWarning: Overwriting tiny_vit_21m_512 in registry with controlnet_aux.segment_anything.modeling.tiny_vit_sam.tiny_vit_21m_512. This is because the name being registered conflicts with an existing name. Please check if this is not expected. | |
| return register_model(fn_wrapper) | |
| You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers | |
| { | |
| "type": "sd_trainer", | |
| "training_folder": "output", | |
| "device": "cuda:0", | |
| "network": { | |
| "type": "lora", | |
| "linear": 16, | |
| "linear_alpha": 16 | |
| }, | |
| "save": { | |
| "dtype": "float16", | |
| "save_every": 500, | |
| "max_step_saves_to_keep": 4, | |
| "push_to_hub": false | |
| }, | |
| "datasets": [ | |
| { | |
| "folder_path": "/workspace/ai-toolkit/images", | |
| "caption_ext": "txt", | |
| "caption_dropout_rate": 0.05, | |
| "shuffle_tokens": false, | |
| "cache_latents_to_disk": true, | |
| "resolution": [ | |
| 512, | |
| 768, | |
| 1024 | |
| ] | |
| } | |
| ], | |
| "train": { | |
| "batch_size": 1, | |
| "steps": 1000, | |
| "gradient_accumulation_steps": 1, | |
| "train_unet": true, | |
| "train_text_encoder": false, | |
| "gradient_checkpointing": true, | |
| "noise_scheduler": "flowmatch", | |
| "optimizer": "adamw8bit", | |
| "lr": 0.0004, | |
| "ema_config": { | |
| "use_ema": true, | |
| "ema_decay": 0.99 | |
| }, | |
| "dtype": "bf16" | |
| }, | |
| "model": { | |
| "name_or_path": "black-forest-labs/FLUX.1-dev", | |
| "is_flux": true, | |
| "quantize": true | |
| }, | |
| "sample": { | |
| "sampler": "flowmatch", | |
| "sample_every": 500, | |
| "width": 1024, | |
| "height": 1024, | |
| "prompts": [ | |
| "Photo of joslora holding a sign that says 'I LOVE PROMPTS!'", | |
| "Professional headshot of joslora in a business suit.", | |
| "A happy pilot joslora of a Boeing 747.", | |
| "A doctor joslora talking to a patient.", | |
| "A chef joslora in the middle of a bustling kitchen, plating a beautifully arranged dish.", | |
| "A young joslora with a big grin, holding a large ice cream cone in front of an old-fashioned ice cream parlor.", | |
| "A person joslora in a tuxedo, looking directly into the camera with a confident smile, standing on a red carpet at a gala event.", | |
| "Person joslora with a bitchin' 80's mullet hairstyle leaning out the window of a pontiac firebird" | |
| ], | |
| "neg": "", | |
| "seed": 42, | |
| "walk_seed": true, | |
| "guidance_scale": 4, | |
| "sample_steps": 20 | |
| }, | |
| "trigger_word": "joslora" | |
| } | |
| Using EMA | |
| ############################################# | |
| # Running job: my_first_flux_lora_v1 | |
| ############################################# | |
| Running 1 process | |
| Loading Flux model | |
| Loading transformer | |
| Quantizing transformer | |
| Loading vae | |
| Loading t5 | |
| Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:04<00:04, 4.08s/it] Downloading shards: 100%|██████████| 2/2 [00:08<00:00, 3.99s/it] Downloading shards: 100%|██████████| 2/2 [00:08<00:00, 4.01s/it] | |
| Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 8.69it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 9.33it/s] | |
| Quantizing T5 | |
| Loading clip | |
| making pipe | |
| preparing | |
| create LoRA network. base dim (rank): 16, alpha: 16 | |
| neuron dropout: p=None, rank dropout: p=None, module dropout: p=None | |
| create LoRA for Text Encoder: 0 modules. | |
| create LoRA for U-Net: 494 modules. | |
| enable LoRA for U-Net | |
| Dataset: /workspace/ai-toolkit/images | |
| - Preprocessing image dimensions | |
| 0%| | 0/8 [00:00<?, ?it/s] 100%|██████████| 8/8 [00:00<00:00, 219.62it/s] | |
| - Found 8 images | |
| Bucket sizes for /workspace/ai-toolkit/images: | |
| 576x384: 8 files | |
| 1 buckets made | |
| Caching latents for /workspace/ai-toolkit/images | |
| - Saving latents to disk | |
| Caching latents to disk: 0%| | 0/8 [00:00<?, ?it/s] Caching latents to disk: 12%|█▎ | 1/8 [00:00<00:02, 2.48it/s] Caching latents to disk: 50%|█████ | 4/8 [00:00<00:00, 8.55it/s] Caching latents to disk: 88%|████████▊ | 7/8 [00:00<00:00, 12.47it/s] Caching latents to disk: 100%|██████████| 8/8 [00:00<00:00, 10.49it/s] | |
| Dataset: /workspace/ai-toolkit/images | |
| - Preprocessing image dimensions | |
| 0%| | 0/8 [00:00<?, ?it/s] 100%|██████████| 8/8 [00:00<00:00, 50610.00it/s] | |
| - Found 8 images | |
| Bucket sizes for /workspace/ai-toolkit/images: | |
| 832x576: 8 files | |
| 1 buckets made | |
| Caching latents for /workspace/ai-toolkit/images | |
| - Saving latents to disk | |
| Caching latents to disk: 0%| | 0/8 [00:00<?, ?it/s] Caching latents to disk: 12%|█▎ | 1/8 [00:00<00:01, 3.65it/s] Caching latents to disk: 38%|███▊ | 3/8 [00:00<00:00, 8.15it/s] Caching latents to disk: 62%|██████▎ | 5/8 [00:00<00:00, 10.02it/s] Caching latents to disk: 88%|████████▊ | 7/8 [00:00<00:00, 11.11it/s] Caching latents to disk: 100%|██████████| 8/8 [00:00<00:00, 9.98it/s] | |
| Dataset: /workspace/ai-toolkit/images | |
| - Preprocessing image dimensions | |
| 0%| | 0/8 [00:00<?, ?it/s] 100%|██████████| 8/8 [00:00<00:00, 63670.65it/s] | |
| - Found 8 images | |
| Bucket sizes for /workspace/ai-toolkit/images: | |
| 1216x832: 8 files | |
| 1 buckets made | |
| Caching latents for /workspace/ai-toolkit/images | |
| - Saving latents to disk | |
| Caching latents to disk: 0%| | 0/8 [00:00<?, ?it/s] Caching latents to disk: 12%|█▎ | 1/8 [00:00<00:00, 8.20it/s] Caching latents to disk: 25%|██▌ | 2/8 [00:00<00:00, 7.99it/s] Caching latents to disk: 38%|███▊ | 3/8 [00:00<00:00, 8.13it/s] Caching latents to disk: 50%|█████ | 4/8 [00:00<00:00, 8.04it/s] Caching latents to disk: 62%|██████▎ | 5/8 [00:00<00:00, 8.06it/s] Caching latents to disk: 75%|███████▌ | 6/8 [00:00<00:00, 8.00it/s] Caching latents to disk: 88%|████████▊ | 7/8 [00:00<00:00, 7.94it/s] Caching latents to disk: 100%|██████████| 8/8 [00:01<00:00, 7.94it/s] Caching latents to disk: 100%|██████████| 8/8 [00:01<00:00, 7.99it/s] | |
| Generating baseline samples before training | |
| Generating Images: 0%| | 0/8 [00:00<?, ?it/s] Generating Images: 12%|█▎ | 1/8 [00:31<03:37, 31.11s/it] Generating Images: 25%|██▌ | 2/8 [00:52<02:30, 25.15s/it] Generating Images: 38%|███▊ | 3/8 [01:13<01:56, 23.26s/it] |