forked from modelscope/facechain
-
Notifications
You must be signed in to change notification settings - Fork 0
/
train_lora.sh
24 lines (21 loc) · 784 Bytes
/
train_lora.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
export MODEL_NAME=$1
export VERSION=$2
export SUB_PATH=$3
export DATASET_NAME=$4
export OUTPUT_DATASET_NAME=$5
export WORK_DIR=$6
accelerate launch facechain/train_text_to_image_lora.py \
--pretrained_model_name_or_path=$MODEL_NAME \
--revision=$VERSION \
--sub_path=$SUB_PATH \
--dataset_name=$DATASET_NAME \
--output_dataset_name=$OUTPUT_DATASET_NAME \
--caption_column="text" \
--resolution=512 --random_flip \
--train_batch_size=1 \
--num_train_epochs=200 --checkpointing_steps=5000 \
--learning_rate=1.5e-04 --lr_scheduler="cosine" --lr_warmup_steps=0 \
--seed=42 \
--output_dir=$WORK_DIR \
--lora_r=4 --lora_alpha=32 \
--lora_text_encoder_r=32 --lora_text_encoder_alpha=32 --resume_from_checkpoint='fromfacecommon'