本文介绍了使用PaddleNLP等工具进行模型训练与推理的流程。先安装paddlenlp等依赖，再分别用dreambooth lora和文生图lora方式训练，设置参数并保存权重。之后可启动visualdl查看训练出图，最后加载训练好的文件，通过相关代码进行推理生成图像。
☞☞☞AI 智能聊天, 问答助手, AI 智能搜索, 免费无限量使用 DeepSeek R1 模型☜☜☜
!pip install -U paddlenlp ppdiffusers safetensors --user
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Requirement already satisfied: paddlenlp in ./.data/webide/pip/lib/python3.7/site-packages (2.5.1) Requirement already satisfied: ppdiffusers in ./.data/webide/pip/lib/python3.7/site-packages (0.11.0) Requirement already satisfied: safetensors in ./.data/webide/pip/lib/python3.7/site-packages (0.2.8) Requirement already satisfied: paddlefsl in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (1.1.0) Requirement already satisfied: sentencepiece in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (0.1.96) Requirement already satisfied: huggingface-hub>=0.11.1 in ./.data/webide/pip/lib/python3.7/site-packages (from paddlenlp) (0.12.0) Requirement already satisfied: seqeval in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (1.2.2) Requirement already satisfied: fastapi in ./.data/webide/pip/lib/python3.7/site-packages (from paddlenlp) (0.91.0) Requirement already satisfied: typer in ./.data/webide/pip/lib/python3.7/site-packages (from paddlenlp) (0.7.0) Requirement already satisfied: multiprocess<=0.70.12.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (0.70.11.1) Requirement already satisfied: jieba in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (0.42.1) Requirement already satisfied: Flask-Babel<3.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (1.0.0) Requirement already satisfied: rich in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (12.6.0) Requirement already satisfied: datasets>=2.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (2.7.0) Requirement already satisfied: paddle2onnx in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (1.0.0) Requirement already satisfied: colorama in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (0.4.4) Requirement already satisfied: visualdl in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (2.4.0) Requirement already satisfied: uvicorn in ./.data/webide/pip/lib/python3.7/site-packages (from paddlenlp) (0.20.0) Requirement already satisfied: tqdm in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (4.64.1) Requirement already satisfied: dill<0.3.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (0.3.3) Requirement already satisfied: colorlog in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from paddlenlp) (4.1.0) Requirement already satisfied: Pillow in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from ppdiffusers) (8.2.0) Requirement already satisfied: regex in ./.data/webide/pip/lib/python3.7/site-packages (from ppdiffusers) (2022.10.31) Requirement already satisfied: ftfy in ./.data/webide/pip/lib/python3.7/site-packages (from ppdiffusers) (6.1.1) Requirement already satisfied: pyyaml>=5.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (5.1.2) Requirement already satisfied: numpy>=1.17 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (1.19.5) Requirement already satisfied: pyarrow>=6.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (10.0.0) Requirement already satisfied: fsspec[http]>=2021.11.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (2022.11.0) Requirement already satisfied: requests>=2.19.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (2.24.0) Requirement already satisfied: xxhash in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (3.1.0) Requirement already satisfied: importlib-metadata in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (4.2.0) Requirement already satisfied: packaging in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (21.3) Requirement already satisfied: aiohttp in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (3.8.3) Requirement already satisfied: responses<0.19 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (0.18.0) Requirement already satisfied: pandas in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from datasets>=2.0.0->paddlenlp) (1.1.5) Requirement already satisfied: Jinja2>=2.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask-Babel<3.0.0->paddlenlp) (3.0.0) Requirement already satisfied: pytz in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask-Babel<3.0.0->paddlenlp) (2019.3) Requirement already satisfied: Babel>=2.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask-Babel<3.0.0->paddlenlp) (2.8.0) Requirement already satisfied: Flask in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask-Babel<3.0.0->paddlenlp) (1.1.1) Requirement already satisfied: filelock in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from huggingface-hub>=0.11.1->paddlenlp) (3.0.12) Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from huggingface-hub>=0.11.1->paddlenlp) (4.3.0) Requirement already satisfied: starlette<0.25.0,>=0.24.0 in ./.data/webide/pip/lib/python3.7/site-packages (from fastapi->paddlenlp) (0.24.0) Requirement already satisfied: pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2 in ./.data/webide/pip/lib/python3.7/site-packages (from fastapi->paddlenlp) (1.10.4) Requirement already satisfied: wcwidth>=0.2.5 in ./.data/webide/pip/lib/python3.7/site-packages (from ftfy->ppdiffusers) (0.2.6) Requirement already satisfied: pygments<3.0.0,>=2.6.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from rich->paddlenlp) (2.13.0) Requirement already satisfied: commonmark<0.10.0,>=0.9.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from rich->paddlenlp) (0.9.1) Requirement already satisfied: scikit-learn>=0.21.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from seqeval->paddlenlp) (0.24.2) Requirement already satisfied: click<9.0.0,>=7.1.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from typer->paddlenlp) (8.0.4) Requirement already satisfied: h21>=0.8 in ./.data/webide/pip/lib/python3.7/site-packages (from uvicorn->paddlenlp) (0.14.0) Requirement already satisfied: matplotlib in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl->paddlenlp) (2.2.3) Requirement already satisfied: six>=1.14.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl->paddlenlp) (1.16.0) Requirement already satisfied: protobuf>=3.11.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl->paddlenlp) (3.20.0) Requirement already satisfied: bce-python-sdk in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl->paddlenlp) (0.8.53) Requirement already satisfied: Werkzeug>=0.15 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask->Flask-Babel<3.0.0->paddlenlp) (0.16.0) Requirement already satisfied: itsdangerous>=0.24 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask->Flask-Babel<3.0.0->paddlenlp) (1.1.0) Requirement already satisfied: attrs>=17.3.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (22.1.0) Requirement already satisfied: aiosignal>=1.1.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (1.2.0) Requirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (6.0.2) Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (4.0.2) Requirement already satisfied: asynctest==0.13.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (0.13.0) Requirement already satisfied: frozenlist>=1.1.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (1.3.0) Requirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (1.7.2) Requirement already satisfied: charset-normalizer<3.0,>=2.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from aiohttp->datasets>=2.0.0->paddlenlp) (2.1.1) Requirement already satisfied: MarkupSafe>=2.0.0rc2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Jinja2>=2.5->Flask-Babel<3.0.0->paddlenlp) (2.0.1) Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from packaging->datasets>=2.0.0->paddlenlp) (3.0.9) Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.19.0->datasets>=2.0.0->paddlenlp) (2019.9.11) Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.19.0->datasets>=2.0.0->paddlenlp) (1.25.11) Requirement already satisfied: chardet<4,>=3.0.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.19.0->datasets>=2.0.0->paddlenlp) (3.0.4) Requirement already satisfied: idna<3,>=2.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.19.0->datasets>=2.0.0->paddlenlp) (2.8) Requirement already satisfied: scipy>=0.19.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from scikit-learn>=0.21.3->seqeval->paddlenlp) (1.6.3) Requirement already satisfied: joblib>=0.11 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from scikit-learn>=0.21.3->seqeval->paddlenlp) (0.14.1) Requirement already satisfied: threadpoolctl>=2.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from scikit-learn>=0.21.3->seqeval->paddlenlp) (2.1.0) Requirement already satisfied: anyio<5,>=3.4.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from starlette<0.25.0,>=0.24.0->fastapi->paddlenlp) (3.6.1) Requirement already satisfied: future>=0.6.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from bce-python-sdk->visualdl->paddlenlp) (0.18.0) Requirement already satisfied: pycryptodome>=3.8.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from bce-python-sdk->visualdl->paddlenlp) (3.9.9) Requirement already satisfied: zipp>=0.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from importlib-metadata->datasets>=2.0.0->paddlenlp) (3.8.1) Requirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl->paddlenlp) (1.1.0) Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl->paddlenlp) (2.8.2) Requirement already satisfied: cycler>=0.10 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl->paddlenlp) (0.10.0) Requirement already satisfied: sniffio>=1.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from anyio<5,>=3.4.0->starlette<0.25.0,>=0.24.0->fastapi->paddlenlp) (1.3.0) Requirement already satisfied: setuptools in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from kiwisolver>=1.0.1->matplotlib->visualdl->paddlenlp) (56.2.0)[notice] A new release of pip available: 22.1.2 -> 23.0.1[notice] To update, run: pip install --upgrade pip
参数解释：
注意：
!python train_dreambooth_lora.py \ --pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5" \ --instance_data_dir="./dogs" \ --output_dir="./dream_booth_lora_outputs" \ --instance_prompt="a photo of sks dog" \ --resolution=512 \ --train_batch_size=1 \ --gradient_accumulation_steps=1 \ --checkpointing_steps=100 \ --learning_rate=1e-4 \ --report_to="visualdl" \ --lr_scheduler="constant" \ --lr_warmup_steps=0 \ --max_train_steps=500 \ --lora_rank=128 \ --validation_prompt="A photo of sks dog in a bucket" \ --validation_epochs=25 \ --validation_guidance_scale=5.0 \ --use_lion False \ --seed=0
[2023-02-23 10:13:09,015] [ WARNING] - You are using a model of type clip_text_model to instantiate a model of type . This is not supported for all configurations of models and can yield errors. W0223 10:13:09.018703 11490 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2 W0223 10:13:09.022612 11490 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2. Train Steps: 20%|███████████████████████▌ | 100/500 [01:12<03:04, 2.16it/s, epoch=0019, step_loss=0.0413]Saved lora weights to ./dream_booth_lora_outputs/checkpoint-100 Train Steps: 40%|███████████████████████████████████████████████▌ | 200/500 [02:25<02:19, 2.15it/s, epoch=0039, step_loss=0.446]Saved lora weights to ./dream_booth_lora_outputs/checkpoint-200 Train Steps: 60%|██████████████████████████████████████████████████████████████████████▏ | 300/500 [03:37<01:33, 2.13it/s, epoch=0059, step_loss=0.00273]Saved lora weights to ./dream_booth_lora_outputs/checkpoint-300 Train Steps: 80%|███████████████████████████████████████████████████████████████████████████████████████████████▏ | 400/500 [04:53<00:47, 2.11it/s, epoch=0079, step_loss=0.275]Saved lora weights to ./dream_booth_lora_outputs/checkpoint-400 Train Steps: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 500/500 [05:44<00:00, 2.14it/s, epoch=0099, step_loss=0.00985]Saved lora weights to ./dream_booth_lora_outputs/checkpoint-500 Saved final lora weights to ./dream_booth_lora_outputs Train Steps: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 500/500 [05:48<00:00, 1.43it/s, epoch=0099, step_loss=0.00985]
--train_data_dir 这个需要放图文对的文件夹，里面是图片和txt。
--image_format 表示 train_data_dir 文件夹内的图片格式，比如png，jpg，jpeg
--use_lion Fasle 表示不使用lion优化器。
!python train_text_to_image_lora.py \ --pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5" \ --output_dir="./text_to_image_lora_outputs3" \ --train_data_dir="mishanwu" \ --image_format="png" \ --resolution=512 \ --train_batch_size=1 \ --gradient_accumulation_steps=1 \ --checkpointing_steps=500 \ --learning_rate=6e-5 \ --report_to="visualdl" \ --lr_scheduler="cosine_with_restarts" \ --lr_warmup_steps=0 \ --max_train_steps=1000 \ --lora_rank=128 \ --validation_prompt="1girl, solo, black_background, looking_at_viewer, parted_lips, tears, brown_eyes" \ --validation_epochs=1 \ --validation_guidance_scale=5.0 \ --use_lion False \ --seed=0
[2023-02-23 11:11:34,386] [ WARNING] - You are using a model of type clip_text_model to instantiate a model of type . This is not supported for all configurations of models and can yield errors. W0223 11:11:34.389356 7327 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 11.2 W0223 11:11:34.392953 7327 gpu_resources.cc:91] device: 0, cuDNN Version: 8.2. Resolving data files: 100%|████████████████| 581/581 [00:00<00:00, 50722.06it/s] Using custom data configuration default-4fb24e511b7f6118 Downloading and preparing dataset imagefolder/default to /home/aistudio/.cache/huggingface/datasets/imagefolder/default-4fb24e511b7f6118/0.0.0/37fbb85cc714a338bea574ac6c7d0b5be5aff46c1862c1989b20e0771199e93f... Downloading data files #0: 0%| | 0/19 [00:00<?, ?obj/s] Downloading data files #3: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #1: 0%| | 0/19 [00:00<?, ?obj/s] Downloading data files #7: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #6: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #4: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #0: 100%|█████████████| 19/19 [00:00<00:00, 3045.97obj/s] Downloading data files #3: 100%|█████████████| 18/18 [00:00<00:00, 2919.24obj/s] Downloading data files #10: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #11: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #1: 100%|█████████████| 19/19 [00:00<00:00, 2589.92obj/s] Downloading data files #4: 100%|█████████████| 18/18 [00:00<00:00, 4414.54obj/s] Downloading data files #7: 100%|█████████████| 18/18 [00:00<00:00, 2468.45obj/s] Downloading data files #6: 100%|█████████████| 18/18 [00:00<00:00, 2742.57obj/s] Downloading data files #2: 100%|█████████████| 19/19 [00:00<00:00, 3250.74obj/s] Downloading data files #12: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #13: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #8: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #10: 100%|████████████| 18/18 [00:00<00:00, 2681.49obj/s] Downloading data files #11: 100%|████████████| 18/18 [00:00<00:00, 2829.21obj/s] Downloading data files #15: 100%|████████████| 18/18 [00:00<00:00, 5331.37obj/s] Downloading data files #5: 100%|█████████████| 18/18 [00:00<00:00, 2877.96obj/s] Downloading data files #13: 100%|████████████| 18/18 [00:00<00:00, 7175.88obj/s] Downloading data files #14: 100%|████████████| 18/18 [00:00<00:00, 9971.93obj/s] Downloading data files #8: 100%|█████████████| 18/18 [00:00<00:00, 6747.47obj/s] Downloading data files #12: 100%|████████████| 18/18 [00:00<00:00, 4458.07obj/s] Downloading data files #9: 100%|█████████████| 18/18 [00:00<00:00, 2966.27obj/s] Downloading data files #2: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #3: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #2: 100%|█████████████| 18/18 [00:00<00:00, 8158.36obj/s] Downloading data files #0: 0%| | 0/19 [00:00<?, ?obj/s] Downloading data files #4: 100%|█████████████| 18/18 [00:00<00:00, 7530.17obj/s] Downloading data files #3: 100%|█████████████| 18/18 [00:00<00:00, 5302.16obj/s] Downloading data files #0: 100%|█████████████| 19/19 [00:00<00:00, 8757.34obj/s] Downloading data files #8: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #9: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #7: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #1: 0%| | 0/19 [00:00<?, ?obj/s] Downloading data files #5: 100%|█████████████| 18/18 [00:00<00:00, 4015.18obj/s] Downloading data files #6: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #8: 100%|█████████████| 18/18 [00:00<00:00, 4227.42obj/s] Downloading data files #7: 100%|█████████████| 18/18 [00:00<00:00, 4900.21obj/s] Downloading data files #6: 100%|█████████████| 18/18 [00:00<00:00, 9804.87obj/s] Downloading data files #11: 100%|████████████| 18/18 [00:00<00:00, 4705.36obj/s] Downloading data files #9: 100%|█████████████| 18/18 [00:00<00:00, 3152.95obj/s] Downloading data files #10: 100%|████████████| 18/18 [00:00<00:00, 6718.65obj/s] Downloading data files #1: 100%|█████████████| 19/19 [00:00<00:00, 3689.60obj/s] Downloading data files #15: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #13: 0%| | 0/18 [00:00<?, ?obj/s] Downloading data files #12: 100%|████████████| 18/18 [00:00<00:00, 6109.20obj/s] Downloading data files #15: 100%|███████████| 18/18 [00:00<00:00, 11363.26obj/s] Downloading data files #13: 100%|███████████| 18/18 [00:00<00:00, 13148.29obj/s] Downloading data files #14: 100%|███████████| 18/18 [00:00<00:00, 16183.81obj/s] Extracting data files #0: 0%| | 0/19 [00:00<?, ?obj/s] Extracting data files #3: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #2: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #1: 0%| | 0/19 [00:00<?, ?obj/s] Extracting data files #7: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #9: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #5: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #6: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #10: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #8: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #4: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #11: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #12: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #13: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #15: 0%| | 0/18 [00:00<?, ?obj/s] Extracting data files #3: 100%|██████████████| 18/18 [00:00<00:00, 1565.14obj/s] Extracting data files #0: 100%|██████████████| 19/19 [00:00<00:00, 1405.75obj/s] Extracting data files #2: 100%|██████████████| 18/18 [00:00<00:00, 1560.64obj/s] Extracting data files #1: 100%|██████████████| 19/19 [00:00<00:00, 1753.47obj/s] Extracting data files #7: 100%|██████████████| 18/18 [00:00<00:00, 1907.66obj/s] Extracting data files #9: 100%|██████████████| 18/18 [00:00<00:00, 1972.81obj/s] Extracting data files #5: 100%|██████████████| 18/18 [00:00<00:00, 2031.58obj/s] Extracting data files #6: 100%|██████████████| 18/18 [00:00<00:00, 2102.00obj/s] Extracting data files #8: 100%|██████████████| 18/18 [00:00<00:00, 2238.62obj/s] Extracting data files #12: 100%|█████████████| 18/18 [00:00<00:00, 2500.58obj/s] Extracting data files #10: 100%|█████████████| 18/18 [00:00<00:00, 1698.37obj/s] Extracting data files #13: 100%|█████████████| 18/18 [00:00<00:00, 2295.87obj/s] Extracting data files #4: 100%|██████████████| 18/18 [00:00<00:00, 1713.79obj/s] Extracting data files #11: 100%|█████████████| 18/18 [00:00<00:00, 1759.56obj/s] Extracting data files #15: 100%|█████████████| 18/18 [00:00<00:00, 2079.59obj/s] Extracting data files #14: 100%|█████████████| 18/18 [00:00<00:00, 1828.07obj/s] Dataset imagefolder downloaded and prepared to /home/aistudio/.cache/huggingface/datasets/imagefolder/default-4fb24e511b7f6118/0.0.0/37fbb85cc714a338bea574ac6c7d0b5be5aff46c1862c1989b20e0771199e93f. Subsequent calls will reuse this data. 100%|████████████████████████████████████████████| 1/1 [00:00<00:00, 256.14it/s] Train Steps: 29%|▎| 290/1000 [02:14<05:22, 2.20it/s, epoch=0000, step_loss=0.1You have disabled the safety checker for <class 'ppdiffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing `safety_checker=None`. Ensure that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. PaddleNLP team, diffusers team and Hugging Face strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling it only for use-cases that involve analyzing network behavior or auditing its results. For more information, please have a look at https://github.com/huggingface/diffusers/pull/254 . Train Steps: 50%|▌| 500/1000 [04:16<03:54, 2.13it/s, epoch=0001, step_loss=0.0Saved lora weights to ./text_to_image_lora_outputs3/checkpoint-500 Train Steps: 58%|▌| 580/1000 [04:55<03:09, 2.22it/s, epoch=0001, step_loss=0.2You have disabled the safety checker for <class 'ppdiffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing `safety_checker=None`. Ensure that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. PaddleNLP team, diffusers team and Hugging Face strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling it only for use-cases that involve analyzing network behavior or auditing its results. For more information, please have a look at https://github.com/huggingface/diffusers/pull/254 . Train Steps: 87%|▊| 870/1000 [07:40<00:58, 2.22it/s, epoch=0002, step_loss=0.0You have disabled the safety checker for <class 'ppdiffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing `safety_checker=None`. Ensure that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. PaddleNLP team, diffusers team and Hugging Face strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling it only for use-cases that involve analyzing network behavior or auditing its results. For more information, please have a look at https://github.com/huggingface/diffusers/pull/254 . Train Steps: 100%|█| 1000/1000 [09:10<00:00, 1.25it/s, epoch=0003, step_loss=0.Saved lora weights to ./text_to_image_lora_outputs3/checkpoint-1000 You have disabled the safety checker for <class 'ppdiffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing `safety_checker=None`. Ensure that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. PaddleNLP team, diffusers team and Hugging Face strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling it only for use-cases that involve analyzing network behavior or auditing its results. For more information, please have a look at https://github.com/huggingface/diffusers/pull/254 . Saved final lora weights to ./text_to_image_lora_outputs3 Train Steps: 100%|█| 1000/1000 [09:44<00:00, 1.71it/s, epoch=0003, step_loss=0.
import lora_helperfrom allinone import StableDiffusionPipelineAllinOnefrom ppdiffusers import DPMSolverMultistepSchedulerimport paddle# 基础模型，需要是paddle版本的权重，未来会加更多的权重pretrained_model_name_or_path = "runwayml/stable-diffusion-v1-5"# 我们加载safetensor版本的权重lora_outputs_path = "9070.safetensors"# 加载之前的模型pipe = StableDiffusionPipelineAllinOne.from_pretrained(pretrained_model_name_or_path, safety_checker=None) pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)# 加载lora权重from IPython.display import clear_output, display clear_output() pipe.apply_lora(lora_outputs_path)
|---------------当前的rank是 128！ |---------------当前的alpha是 128.0！ Loading lora_weights successfully!
import lora_helperfrom allinone import StableDiffusionPipelineAllinOnefrom ppdiffusers import DPMSolverMultistepScheduler prompt = "A photo of sks dog in a bucket"negative_prompt = ""guidance_scale = 8num_inference_steps = 25height = 512width = 512img = pipe(prompt, negative_prompt=negative_prompt, guidance_scale=guidance_scale, height=height, width=width, num_inference_steps=num_inference_steps).images[0] display(img) display(img.argument)
0%| | 0/25 [00:00<?, ?it/s]
<PIL.Image.Image image mode=RGB size=512x512 at 0x7FCA25135290>
{'prompt': 'A photo of sks dog in a bucket',
'negative_prompt': '',
'height': 512,
'width': 512,
'num_inference_steps': 25,
'guidance_scale': 8,
'num_images_per_prompt': 1,
'eta': 0.0,
'seed': 3574959348,
'latents': None,
'max_embeddings_multiples': 1,
'no_boseos_middle': False,
'skip_parsing': False,
'skip_weighting': False,
'epoch_time': 1676862593.5281246}以上就是使用Lora技术进行Dreambooth训练【抢先体验版】的详细内容，更多请关注php中文网其它相关文章！
每个人都需要一台速度更快、更稳定的 PC。随着时间的推移，垃圾文件、旧注册表数据和不必要的后台进程会占用资源并降低性能。幸运的是，许多工具可以让 Windows 保持平稳运行。
Copyright 2014-2025 https://www.php.cn/ All Rights Reserved | php.cn | 湘ICP备2023035733号
955
收藏
