github
/
ChatGLM-6B
mirror of https://github.com/THUDM/ChatGLM-6B
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Add validation file name 

Use full prediction
pull/313/head
duzx16 2023-03-31 20:15:35 +08:00
parent 08d880141d
commit 73f4fe1ffe
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ptuning/evaluate.sh
View File

@ -4,6 +4,7 @@ STEP=3000
CUDA_VISIBLE_DEVICES=0 python3 main.py \ CUDA_VISIBLE_DEVICES=0 python3 main.py \
--do_predict \ --do_predict \
--validation_file AdvertiseGen/dev.json \
--test_file AdvertiseGen/dev.json \ --test_file AdvertiseGen/dev.json \
--overwrite_cache \ --overwrite_cache \
--prompt_column content \ --prompt_column content \
@ -15,6 +16,5 @@ CUDA_VISIBLE_DEVICES=0 python3 main.py \
--max_target_length 64 \ --max_target_length 64 \
--per_device_eval_batch_size 1 \ --per_device_eval_batch_size 1 \
--predict_with_generate \ --predict_with_generate \
--max_predict_samples 10 \
--pre_seq_len $PRE_SEQ_LEN \ --pre_seq_len $PRE_SEQ_LEN \
--quantization_bit 4 --quantization_bit 4

2
ptuning/train.sh
View File

@ -9,7 +9,7 @@ CUDA_VISIBLE_DEVICES=0 python3 main.py \
--response_column summary \ --response_column summary \
--overwrite_cache \ --overwrite_cache \
--model_name_or_path THUDM/chatglm-6b \ --model_name_or_path THUDM/chatglm-6b \
--output_dir output/adgen-chatglm-6b-pt-$PRE_SEQ_LEN-$LR-dev \ --output_dir output/adgen-chatglm-6b-pt-$PRE_SEQ_LEN-$LR \
--overwrite_output_dir \ --overwrite_output_dir \
--max_source_length 64 \ --max_source_length 64 \
--max_target_length 64 \ --max_target_length 64 \