cocktailpeanut / fluxgym

Dead simple FLUX LoRA training UI with LOW VRAM support
MIT License
1.35k stars 110 forks source link

TRAINING COMPLETE Error code 1 loading t5xxl_fp16 #201

Open T4CAi opened 1 month ago

T4CAi commented 1 month ago

Hey everyone, I know I'm not the only one with this issue but still have no solution so I will try a new thread.... After a few seconds after I start the training, I get "TRAINING COMPLETE" with an error.... I don't have any special characters in the captions, I have 8Gb Vram (but saw others for whom this works)

If anyone could help I would be very grateful 🙏 Thanks for reading

INFO Loaded CLIP: <All keys matched flux_utils.py:261 [2024-10-18 15:38:40] [INFO] successfully> [2024-10-18 15:38:40] [INFO] INFO Loading state dict from flux_utils.py:306 [2024-10-18 15:38:40] [INFO] C:\pinokio\api\fluxgym.git\model [2024-10-18 15:38:40] [INFO] s\clip\t5xxl_fp16.safetensors [2024-10-18 15:38:44] [INFO] Traceback (most recent call last): [2024-10-18 15:38:44] [INFO] File "C:\pinokio\bin\miniconda\lib\runpy.py", line 196, in _run_module_as_main [2024-10-18 15:38:44] [INFO] return _run_code(code, main_globals, None, [2024-10-18 15:38:44] [INFO] File "C:\pinokio\bin\miniconda\lib\runpy.py", line 86, in _run_code [2024-10-18 15:38:44] [INFO] exec(code, run_globals) [2024-10-18 15:38:44] [INFO] File "C:\pinokio\api\fluxgym.git\env\Scripts\accelerate.exe__main__.py", line 7, in [2024-10-18 15:38:44] [INFO] sys.exit(main()) [2024-10-18 15:38:44] [INFO] File "C:\pinokio\api\fluxgym.git\env\lib\site-packages\accelerate\commands\accelerate_cli.py", line 48, in main [2024-10-18 15:38:44] [INFO] args.func(args) [2024-10-18 15:38:44] [INFO] File "C:\pinokio\api\fluxgym.git\env\lib\site-packages\accelerate\commands\launch.py", line 1106, in launch_command [2024-10-18 15:38:44] [INFO] simple_launcher(args) [2024-10-18 15:38:44] [INFO] File "C:\pinokio\api\fluxgym.git\env\lib\site-packages\accelerate\commands\launch.py", line 704, in simple_launcher [2024-10-18 15:38:44] [INFO] raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) [2024-10-18 15:38:44] [INFO] subprocess.CalledProcessError: Command '['C:\pinokio\api\fluxgym.git\env\Scripts\python.exe', 'sd-scripts/flux_train_network.py', '--pretrained_model_name_or_path', 'C:\pinokio\api\fluxgym.git\models\unet\flux1-dev.sft', '--clip_l', 'C:\pinokio\api\fluxgym.git\models\clip\clip_l.safetensors', '--t5xxl', 'C:\pinokio\api\fluxgym.git\models\clip\t5xxl_fp16.safetensors', '--ae', 'C:\pinokio\api\fluxgym.git\models\vae\ae.sft', '--cache_latents_to_disk', '--save_model_as', 'safetensors', '--sdpa', '--persistent_data_loader_workers', '--max_data_loader_n_workers', '2', '--seed', '42', '--gradient_checkpointing', '--mixed_precision', 'bf16', '--save_precision', 'bf16', '--network_module', 'networks.lora_flux', '--network_dim', '4', '--optimizer_type', 'adafactor', '--optimizer_args', 'relative_step=False', 'scale_parameter=False', 'warmup_init=False', '--split_mode', '--network_args', 'train_blocks=single', '--lr_scheduler', 'constant_with_warmup', '--max_grad_norm', '0.0', '--sample_prompts=C:\pinokio\api\fluxgym.git\outputs\tchardj\sample_prompts.txt', '--sample_every_n_steps=380', '--learning_rate', '8e-4', '--cache_text_encoder_outputs', '--cache_text_encoder_outputs_to_disk', '--fp8_base', '--highvram', '--max_train_epochs', '16', '--save_every_n_epochs', '4', '--dataset_config', 'C:\pinokio\api\fluxgym.git\outputs\tchardj\dataset.toml', '--output_dir', 'C:\pinokio\api\fluxgym.git\outputs\tchardj', '--output_name', 'tchardj', '--timestep_sampling', 'shift', '--discrete_flow_shift', '3.1582', '--model_prediction_type', 'raw', '--guidance_scale', '1', '--loss_type', 'l2', '--cache_latents']' returned non-zero exit status 3221225477. [2024-10-18 15:38:45] [ERROR] Command exited with code 1 [2024-10-18 15:38:45] [INFO] Runner: