“AIGC Stable Diffusion文生图Lora模型微调实现虚拟上装”Lora微调训练问题

在执行Lora模型训练这一步的时候:执行如下命令,设置num_train_epochs为200,进行lora模型的训练:
! export MODEL_NAME="chilloutmix-ni" && \
export DATASET_NAME="cloth_train_example" && \
accelerate launch --mixed_precision="fp16" train_text_to_image_lora.py \
--pretrained_model_name_or_path=$MODEL_NAME \
--dataset_name=$DATASET_NAME --caption_column="text" \
--width=640 --height=768 --random_flip \
--train_batch_size=1 \
--num_train_epochs=200 --checkpointing_steps=5000 \
--learning_rate=1e-04 --lr_scheduler="constant" --lr_warmup_steps=0 \
--seed=42 \
--output_dir="cloth-model-lora" \
--validation_prompt="cloth1" --validation_epochs=100

出现问题:

╭───────────────────── Traceback (most recent call last) ──────────────────────╮
│ /usr/local/lib/python3.10/dist-packages/diffusers/configuration_utils.py:337 │
│ in load_config │
│ │
│ 334 │ │ else: │
│ 335 │ │ │ try: │
│ 336 │ │ │ │ # Load from URL or cache if already cached │
│ ❱ 337 │ │ │ │ config_file = hf_hub_download( │
│ 338 │ │ │ │ │ pretrained_model_name_or_path, │
│ 339 │ │ │ │ │ filename=cls.config_name, │
│ 340 │ │ │ │ │ cache_dir=cache_dir, │
│ │
│ /usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_validators.py │
│ :120 in _inner_fn │
│ │
│ 117 │ │ if check_use_auth_token: │
│ 118 │ │ │ kwargs = smoothly_deprecate_use_auth_token(fn_name=fn.__na │
│ 119 │ │ │
│ ❱ 120 │ │ return fn(args, *kwargs) │
│ 121 │ │
│ 122 │ return _inner_fn # type: ignore │
│ 123 │
│ │
│ /usr/local/lib/python3.10/dist-packages/huggingface_hub/file_download.py:125 │
│ 3 in hf_hub_download │
│ │
│ 1250 │ │ │ │ " and downloads online, set 'local_files_only' to Fal │
│ 1251 │ │ │ ) │
│ 1252 │ │ else: │
│ ❱ 1253 │ │ │ raise LocalEntryNotFoundError( │
│ 1254 │ │ │ │ "Connection error, and we cannot find the requested f │
│ 1255 │ │ │ │ " the disk cache. Please try again or make sure your │
│ 1256 │ │ │ │ " connection is on." │
╰──────────────────────────────────────────────────────────────────────────────╯
LocalEntryNotFoundError: Connection error, and we cannot find the requested
files in the disk cache. Please try again or make sure your Internet connection
is on.

During handling of the above exception, another exception occurred:

╭───────────────────── Traceback (most recent call last) ──────────────────────╮
│ /mnt/workspace/train_text_to_image_lora.py:896 in │
│ │
│ 893 │
│ 894 │
│ 895 if name == "main": │
│ ❱ 896 │ main() │
│ 897 │
│ │
│ /mnt/workspace/train_text_to_image_lora.py:440 in main │
│ │
│ 437 │ │ │ os.makedirs(args.output_dir, exist_ok=True) │
│ 438 │ │
│ 439 │ # Load scheduler, tokenizer and models. │
│ ❱ 440 │ noise_scheduler = DDPMScheduler.from_pretrained(args.pretrained_mo │
│ 441 │ tokenizer = CLIPTokenizer.from_pretrained( │
│ 442 │ │ args.pretrained_model_name_or_path, subfolder="tokenizer", rev │
│ 443 │ ) │
│ │
│ /usr/local/lib/python3.10/dist-packages/diffusers/schedulers/scheduling_util │
│ s.py:139 in from_pretrained │
│ │
│ 136 │ │ │
│ 137 │ │ │
│ 138 │ │ """ │
│ ❱ 139 │ │ config, kwargs, commit_hash = cls.load_config( │
│ 140 │ │ │ pretrained_model_name_or_path=pretrained_model_name_or_pat │
│ 141 │ │ │ subfolder=subfolder, │
│ 142 │ │ │ return_unused_kwargs=True, │
│ │
│ /usr/local/lib/python3.10/dist-packages/diffusers/configuration_utils.py:364 │
│ in load_config │
│ │
│ 361 │ │ │ │ │ f" 'https://huggingface.co/{pretrained_model_name_
│ 362 │ │ │ │ ) │
│ 363 │ │ │ except EntryNotFoundError: │
│ ❱ 364 │ │ │ │ raise EnvironmentError( │
│ 365 │ │ │ │ │ f"{pretrained_model_name_or_path} does not appear │
│ 366 │ │ │ │ ) │
│ 367 │ │ │ except HTTPError as err: │
╰──────────────────────────────────────────────────────────────────────────────╯
OSError: chilloutmix-ni does not appear to have a file named
scheduler_config.json.
[02:18:28] ERROR failed (exitcode: 1) local_rank: 0 (pid: 502) of ]8;id=222292;file:///usr/local/lib/python3.10/dist-packages/torch/distributed/elastic/multiprocessing/api.py\api.py]8;;:]8;id=930720;file:///usr/local/lib/python3.10/dist-packages/torch/distributed/elastic/multiprocessing/api.py#673\673]8;;

binary: /usr/bin/python3

展开
收起
5rvcrq3dien7u 2023-12-22 11:09:29 266 分享 版权
1 条回答
写回答
取消 提交回答
  • 问题解决啦!我自己的问题,因为隔天重启实例,没有从第一步做起,直接接着做,所以一直重现昨天的错误。

    2023-12-22 13:01:17
    赞同 展开评论

人工智能平台 PAI(Platform for AI,原机器学习平台PAI)是面向开发者和企业的机器学习/深度学习工程平台,提供包含数据标注、模型构建、模型训练、模型部署、推理优化在内的AI开发全链路服务,内置140+种优化算法,具备丰富的行业场景插件,为用户提供低门槛、高性能的云原生AI工程化能力。

还有其他疑问?
咨询AI助理