mirror of https://github.com/hpcaitech/ColossalAI
25 lines
690 B
Bash
25 lines
690 B
Bash
#!/bin/bash
|
|
set -xe
|
|
pip install -r requirements.txt
|
|
|
|
HF_DATASETS_OFFLINE=1
|
|
TRANSFORMERS_OFFLINE=1
|
|
DIFFUSERS_OFFLINE=1
|
|
|
|
# "torch_ddp" "torch_ddp_fp16"
|
|
for plugin in "low_level_zero" "gemini"; do
|
|
torchrun --nproc_per_node 8 --standalone train_dreambooth_colossalai.py \
|
|
--pretrained_model_name_or_path="/data/dreambooth/diffuser/stable-diffusion-v1-4" \
|
|
--instance_data_dir="/data/dreambooth/Teyvat/data" \
|
|
--output_dir="./weight_output" \
|
|
--instance_prompt="a picture of a dog" \
|
|
--resolution=512 \
|
|
--plugin=$plugin \
|
|
--train_batch_size=1 \
|
|
--learning_rate=5e-6 \
|
|
--lr_scheduler="constant" \
|
|
--lr_warmup_steps=0 \
|
|
--num_class_images=200 \
|
|
--placement="cuda"
|
|
done
|