File size: 691 Bytes
fa6856c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
TOT_CUDA="0,1"
CUDAs=(${TOT_CUDA//,/ })
CUDA_NUM=${#CUDAs[@]}
PORT="12345"

DATA_PATH="./sample/instruct/data_sample.json" #"../dataset/instruction/guanaco_non_chat_mini_52K-utf8.json"
OUTPUT_PATH="../lora-Vicuna-new"
MODEL_PATH="../llama-13b-hf/"
lora_checkpoint="../Chinese-Vicuna-lora-13b-belle-and-guanaco"
from_data_beginning=True # False
TEST_SIZE=1000

CUDA_VISIBLE_DEVICES=${TOT_CUDA} torchrun --nproc_per_node=$CUDA_NUM --master_port=$PORT finetune.py \
--data_path $DATA_PATH \
--output_path $OUTPUT_PATH \
--model_path $MODEL_PATH \
--eval_steps 200 \
--save_steps 200 \
--test_size $TEST_SIZE \
--resume_from_checkpoint $lora_checkpoint \
--ignore_data_skip $from_data_beginning