AdGPT / README.md
goodmodeler's picture
add train lora
686c304
|
raw
history blame
1.67 kB
metadata
title: ZeroGPU
emoji: 🖼
colorFrom: purple
colorTo: red
sdk: gradio
sdk_version: 5.25.2
app_file: app.py
pinned: false
license: apache-2.0

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

commands:

pip install git+https://github.com/huggingface/diffusers

accelerate launch
--deepspeed_config_file ds_config.json
diffusers/examples/dreambooth/train_dreambooth.py
--pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5"
--instance_data_dir="./nyc_ads_dataset"
--instance_prompt="a photo of an urbanad nyc"
--output_dir="./nyc-ad-model"
--resolution=100
--train_batch_size=1
--gradient_accumulation_steps=1
--gradient_checkpointing
--learning_rate=5e-6
--lr_scheduler="constant"
--lr_warmup_steps=0
--max_train_steps=400
--mixed_precision="fp16"
--checkpointing_steps=100
--checkpoints_total_limit=1
--report_to="tensorboard"
--logging_dir="./nyc-ad-model/logs"

fine tune a trained model: --pretrained_model_name_or_path="./nyc-ad-model/checkpoint-400" \

export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True

import torch torch.cuda.empty_cache() torch.cuda.reset_peak_memory_stats()

7/12

1 Fine‑tune image model LoRA+QLoRA

accelerate launch --deepspeed_config_file=ds_config_zero3.json train_lora.py python train_lora.py

2 SFT 语言模型

python sft_train.py

3 Build RAG index

python build_embeddings.py

4 (可选) 收集偏好 → 训练 reward model

python reward_model.py

5 PPO RLHF 微调

python ppo_tune.py

6 Inference with RAG

python rag_infer.py