Compare commits
3 Commits
24f58972f3
...
dfbc8a2810
| Author | SHA1 | Date |
|---|---|---|
wql
|
dfbc8a2810 | |
|
wql
|
e434505c6b | |
|
wql
|
47b4cd3a54 |
96
batch_run.sh
96
batch_run.sh
|
|
@ -1,58 +1,58 @@
|
|||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_1.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_1_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_2.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_2_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_3.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_3_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_1_single.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_1_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_2_single.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_2_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_3_single.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_3_single_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_1.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_1_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_2.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_2_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_3.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_3_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_1_single.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_1_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_2_single.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_2_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Llama2-7B/llama2_lora_sft_3_single.yaml | tee results/lora_sft_2/Llama2-7B/llama2_lora_sft_3_single_log.txt
|
||||
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1_single.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2_single.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3_single.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3_single_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1_single.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_1_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2_single.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_2_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3_single.yaml | tee results/lora_sft_2/Baichuan2-7B/Baichuan2_lora_sft_3_single_log.txt
|
||||
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1_single.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2_single.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3_single.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3_single_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1_single.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_1_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2_single.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_2_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3_single.yaml | tee results/lora_sft_2/ChatGLM2-6B/ChatGLM2_lora_sft_3_single_log.txt
|
||||
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1_single.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2_single.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2_single_step500_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3_single.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3_single_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1_single.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_1_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2_single.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_2_single_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3_single.yaml | tee results/lora_sft_2/Qwen-7B/Qwen_lora_sft_3_single_log.txt
|
||||
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_1.yaml | tee results/inference/Llama2-7B/llama2_predict_1_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_2.yaml | tee results/inference/Llama2-7B/llama2_predict_2_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_3.yaml | tee results/inference/Llama2-7B/llama2_predict_3_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_1_single.yaml | tee results/inference/Llama2-7B/llama2_predict_1_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_2_single.yaml | tee results/inference/Llama2-7B/llama2_predict_2_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_3_single.yaml | tee results/inference/Llama2-7B/llama2_predict_3_single_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_1.yaml | tee results/inference/Llama2-7B/llama2_predict_1_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_2.yaml | tee results/inference/Llama2-7B/llama2_predict_2_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_3.yaml | tee results/inference/Llama2-7B/llama2_predict_3_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_1_single.yaml | tee results/inference/Llama2-7B/llama2_predict_1_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_2_single.yaml | tee results/inference/Llama2-7B/llama2_predict_2_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_3_single.yaml | tee results/inference/Llama2-7B/llama2_predict_3_single_log.txt
|
||||
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_1.yaml | tee results/inference/Qwen-7B/Qwen_predict_1_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_2.yaml | tee results/inference/Qwen-7B/Qwen_predict_2_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_3.yaml | tee results/inference/Qwen-7B/Qwen_predict_3_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_1_single.yaml | tee results/inference/Qwen-7B/Qwen_predict_1_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_2_single.yaml | tee results/inference/Qwen-7B/Qwen_predict_2_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_3_single.yaml | tee results/inference/Qwen-7B/Qwen_predict_3_single_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_1.yaml | tee results/inference/Qwen-7B/Qwen_predict_1_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_2.yaml | tee results/inference/Qwen-7B/Qwen_predict_2_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_3.yaml | tee results/inference/Qwen-7B/Qwen_predict_3_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_1_single.yaml | tee results/inference/Qwen-7B/Qwen_predict_1_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_2_single.yaml | tee results/inference/Qwen-7B/Qwen_predict_2_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Qwen-7B/Qwen_predict_3_single.yaml | tee results/inference/Qwen-7B/Qwen_predict_3_single_log.txt
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_1.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_1_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_2.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_2_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_3.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_3_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_1_single.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_1_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_2_single.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_2_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_3_single.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_3_single_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_1.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_1_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_2.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_2_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_3.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_3_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_1_single.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_1_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_2_single.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_2_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/ChatGLM2-6B/ChatGLM2_predict_3_single.yaml | tee results/inference/ChatGLM2-6B/ChatGLM2_predict_3_single_log.txt
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_1.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_1_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_2.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_2_step500_log.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_3.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_3_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_1_single.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_1_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_2_single.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_2_single_step500_log.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_3_single.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_3_single_step500_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_1.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_1_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_2.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_2_log.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_3.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_3_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_1_single.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_1_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_2_single.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_2_single_log.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Baichuan2-7B/Baichuan2_predict_3_single.yaml | tee results/inference/Baichuan2-7B/Baichuan2_predict_3_single_log.txt
|
||||
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Baichuan2-7B/Baichuan2_predict_1
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Baichuan2-7B/Baichuan2_predict_1_single
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Baichuan2-7B/Baichuan2_predict_2
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Baichuan2-7B/Baichuan2_predict_2_single
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Baichuan2-7B/Baichuan2_predict_3
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Baichuan2-7B/Baichuan2_predict_3_single
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ../../llm/chatglm/data
|
||||
|
||||
### method
|
||||
do_predict: true
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/ChatGLM2-6B/ChatGLM2_predict_1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ../../llm/chatglm/data
|
||||
|
||||
### method
|
||||
do_predict: true
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/ChatGLM2-6B/ChatGLM2_predict_1_single
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ../../llm/chatglm/data
|
||||
|
||||
### method
|
||||
do_predict: true
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/ChatGLM2-6B/ChatGLM2_predict_2
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ../../llm/chatglm/data
|
||||
|
||||
### method
|
||||
do_predict: true
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/ChatGLM2-6B/ChatGLM2_predict_2_single
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ../../llm/chatglm/data
|
||||
|
||||
### method
|
||||
do_predict: true
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/ChatGLM2-6B/ChatGLM2_predict_3
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ../../llm/chatglm/data
|
||||
|
||||
### method
|
||||
do_predict: true
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/ChatGLM2-6B/ChatGLM2_predict_3_single
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.4764,
|
||||
"predict_samples_per_second": 0.788,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.4764,
|
||||
"predict_samples_per_second": 0.788,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
|
|
@ -1,58 +0,0 @@
|
|||
08/18/2024 17:34:57 - INFO - llamafactory.cli - Initializing distributed tasks at: 127.0.0.1:23898
|
||||
08/18/2024 17:35:04 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:04 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:05 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:05 - INFO - llamafactory.hparams.parser - Process rank: 4, device: cuda:4, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:05 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:05 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:05 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:05 - INFO - llamafactory.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:05 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:05 - INFO - llamafactory.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:05 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:05 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:05 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:35:05 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:35:06 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:35:16 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:35:16 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:35:16 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:35:16 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:35:16 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:35:16 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
eval example:
|
||||
input_ids:
|
||||
[1, 518, 25580, 29962, 29871, 30982, 31695, 31863, 31577, 30210, 30457, 30502, 31302, 30858, 30267, 518, 29914, 25580, 29962]
|
||||
inputs:
|
||||
<s> [INST] 保持健康的三个提示。 [/INST]
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:35:34 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:36:13 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:13 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:36:13 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:13 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:36:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
***** predict metrics *****
|
||||
predict_bleu-4 = 1.4552
|
||||
predict_model_preparation_time = 0.0031
|
||||
predict_rouge-1 = 11.0984
|
||||
predict_rouge-2 = 3.0446
|
||||
predict_rouge-l = 5.0312
|
||||
predict_runtime = 0:01:03.50
|
||||
predict_samples_per_second = 0.787
|
||||
predict_steps_per_second = 0.063
|
||||
08/18/2024 17:37:18 - INFO - llamafactory.train.sft.trainer - Saving prediction results to ./results/inference/Llama2-7B/llama2_predict_1/generated_predictions.jsonl
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Llama2-7B/llama2_predict_1
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.0031,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.5091,
|
||||
"predict_samples_per_second": 0.787,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.0031,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.5091,
|
||||
"predict_samples_per_second": 0.787,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
|
|
@ -1,20 +0,0 @@
|
|||
08/18/2024 17:42:59 - INFO - llamafactory.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, compute dtype: None
|
||||
08/18/2024 17:43:00 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
eval example:
|
||||
input_ids:
|
||||
[1, 518, 25580, 29962, 29871, 30982, 31695, 31863, 31577, 30210, 30457, 30502, 31302, 30858, 30267, 518, 29914, 25580, 29962]
|
||||
inputs:
|
||||
<s> [INST] 保持健康的三个提示。 [/INST]
|
||||
08/18/2024 17:43:05 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:43:34 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:43:34 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
***** predict metrics *****
|
||||
predict_bleu-4 = 1.2952
|
||||
predict_model_preparation_time = 0.0029
|
||||
predict_rouge-1 = 12.8381
|
||||
predict_rouge-2 = 3.2551
|
||||
predict_rouge-l = 5.1021
|
||||
predict_runtime = 0:06:01.55
|
||||
predict_samples_per_second = 0.138
|
||||
predict_steps_per_second = 0.069
|
||||
08/18/2024 17:49:36 - INFO - llamafactory.train.sft.trainer - Saving prediction results to ./results/inference/Llama2-7B/llama2_predict_1_single/generated_predictions.jsonl
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Llama2-7B/llama2_predict_1_single
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.2952020000000002,
|
||||
"predict_model_preparation_time": 0.0029,
|
||||
"predict_rouge-1": 12.838101999999997,
|
||||
"predict_rouge-2": 3.255076,
|
||||
"predict_rouge-l": 5.102102,
|
||||
"predict_runtime": 361.551,
|
||||
"predict_samples_per_second": 0.138,
|
||||
"predict_steps_per_second": 0.069
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.2952020000000002,
|
||||
"predict_model_preparation_time": 0.0029,
|
||||
"predict_rouge-1": 12.838101999999997,
|
||||
"predict_rouge-2": 3.255076,
|
||||
"predict_rouge-l": 5.102102,
|
||||
"predict_runtime": 361.551,
|
||||
"predict_samples_per_second": 0.138,
|
||||
"predict_steps_per_second": 0.069
|
||||
}
|
||||
|
|
@ -1,5 +0,0 @@
|
|||
{"current_steps": 5, "total_steps": 25, "percentage": 20.0, "elapsed_time": "0:00:52", "remaining_time": "0:03:29"}
|
||||
{"current_steps": 10, "total_steps": 25, "percentage": 40.0, "elapsed_time": "0:02:05", "remaining_time": "0:03:07"}
|
||||
{"current_steps": 15, "total_steps": 25, "percentage": 60.0, "elapsed_time": "0:03:18", "remaining_time": "0:02:12"}
|
||||
{"current_steps": 20, "total_steps": 25, "percentage": 80.0, "elapsed_time": "0:04:31", "remaining_time": "0:01:07"}
|
||||
{"current_steps": 25, "total_steps": 25, "percentage": 100.0, "elapsed_time": "0:05:44", "remaining_time": "0:00:00"}
|
||||
|
|
@ -1,58 +0,0 @@
|
|||
08/18/2024 17:37:25 - INFO - llamafactory.cli - Initializing distributed tasks at: 127.0.0.1:28639
|
||||
08/18/2024 17:37:31 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:31 - INFO - llamafactory.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:32 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:32 - INFO - llamafactory.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:32 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:32 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:33 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:33 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:33 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:33 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:33 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:33 - INFO - llamafactory.hparams.parser - Process rank: 4, device: cuda:4, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:33 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:37:33 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:37:33 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:38:38 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:38:38 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:38:38 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:38:38 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:38:38 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:38:38 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
eval example:
|
||||
input_ids:
|
||||
[1, 518, 25580, 29962, 29871, 30982, 31695, 31863, 31577, 30210, 30457, 30502, 31302, 30858, 30267, 518, 29914, 25580, 29962]
|
||||
inputs:
|
||||
<s> [INST] 保持健康的三个提示。 [/INST]
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:38:44 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:39:22 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:22 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:39:23 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:23 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:39:25 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
***** predict metrics *****
|
||||
predict_bleu-4 = 1.4552
|
||||
predict_model_preparation_time = 0.003
|
||||
predict_rouge-1 = 11.0984
|
||||
predict_rouge-2 = 3.0446
|
||||
predict_rouge-l = 5.0312
|
||||
predict_runtime = 0:01:03.40
|
||||
predict_samples_per_second = 0.789
|
||||
predict_steps_per_second = 0.063
|
||||
08/18/2024 17:40:28 - INFO - llamafactory.train.sft.trainer - Saving prediction results to ./results/inference/Llama2-7B/llama2_predict_2/generated_predictions.jsonl
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Llama2-7B/llama2_predict_2
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.4089,
|
||||
"predict_samples_per_second": 0.789,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.4089,
|
||||
"predict_samples_per_second": 0.789,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
|
|
@ -1,20 +0,0 @@
|
|||
08/18/2024 17:49:41 - INFO - llamafactory.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, compute dtype: None
|
||||
08/18/2024 17:49:42 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
eval example:
|
||||
input_ids:
|
||||
[1, 518, 25580, 29962, 29871, 30982, 31695, 31863, 31577, 30210, 30457, 30502, 31302, 30858, 30267, 518, 29914, 25580, 29962]
|
||||
inputs:
|
||||
<s> [INST] 保持健康的三个提示。 [/INST]
|
||||
08/18/2024 17:49:47 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:50:16 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:50:16 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
***** predict metrics *****
|
||||
predict_bleu-4 = 1.2952
|
||||
predict_model_preparation_time = 0.003
|
||||
predict_rouge-1 = 12.8381
|
||||
predict_rouge-2 = 3.2551
|
||||
predict_rouge-l = 5.1021
|
||||
predict_runtime = 0:06:01.42
|
||||
predict_samples_per_second = 0.138
|
||||
predict_steps_per_second = 0.069
|
||||
08/18/2024 17:56:18 - INFO - llamafactory.train.sft.trainer - Saving prediction results to ./results/inference/Llama2-7B/llama2_predict_2_single/generated_predictions.jsonl
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Llama2-7B/llama2_predict_2_single
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.2952020000000002,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 12.838101999999997,
|
||||
"predict_rouge-2": 3.255076,
|
||||
"predict_rouge-l": 5.102102,
|
||||
"predict_runtime": 361.4262,
|
||||
"predict_samples_per_second": 0.138,
|
||||
"predict_steps_per_second": 0.069
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.2952020000000002,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 12.838101999999997,
|
||||
"predict_rouge-2": 3.255076,
|
||||
"predict_rouge-l": 5.102102,
|
||||
"predict_runtime": 361.4262,
|
||||
"predict_samples_per_second": 0.138,
|
||||
"predict_steps_per_second": 0.069
|
||||
}
|
||||
|
|
@ -1,5 +0,0 @@
|
|||
{"current_steps": 5, "total_steps": 25, "percentage": 20.0, "elapsed_time": "0:00:52", "remaining_time": "0:03:28"}
|
||||
{"current_steps": 10, "total_steps": 25, "percentage": 40.0, "elapsed_time": "0:02:05", "remaining_time": "0:03:07"}
|
||||
{"current_steps": 15, "total_steps": 25, "percentage": 60.0, "elapsed_time": "0:03:17", "remaining_time": "0:02:11"}
|
||||
{"current_steps": 20, "total_steps": 25, "percentage": 80.0, "elapsed_time": "0:04:31", "remaining_time": "0:01:07"}
|
||||
{"current_steps": 25, "total_steps": 25, "percentage": 100.0, "elapsed_time": "0:05:44", "remaining_time": "0:00:00"}
|
||||
|
|
@ -1,58 +0,0 @@
|
|||
08/18/2024 17:40:35 - INFO - llamafactory.cli - Initializing distributed tasks at: 127.0.0.1:24151
|
||||
08/18/2024 17:40:42 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:42 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:43 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:43 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:43 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:43 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:43 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:43 - INFO - llamafactory.hparams.parser - Process rank: 4, device: cuda:4, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:43 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:43 - INFO - llamafactory.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:43 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:43 - INFO - llamafactory.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:43 - WARNING - llamafactory.hparams.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
|
||||
08/18/2024 17:40:43 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: None
|
||||
08/18/2024 17:40:44 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:40:54 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:40:54 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:40:54 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:40:54 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:40:54 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
08/18/2024 17:40:54 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
eval example:
|
||||
input_ids:
|
||||
[1, 518, 25580, 29962, 29871, 30982, 31695, 31863, 31577, 30210, 30457, 30502, 31302, 30858, 30267, 518, 29914, 25580, 29962]
|
||||
inputs:
|
||||
<s> [INST] 保持健康的三个提示。 [/INST]
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:09 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:48 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
08/18/2024 17:41:49 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:41:49 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
***** predict metrics *****
|
||||
predict_bleu-4 = 1.4552
|
||||
predict_model_preparation_time = 0.0031
|
||||
predict_rouge-1 = 11.0984
|
||||
predict_rouge-2 = 3.0446
|
||||
predict_rouge-l = 5.0312
|
||||
predict_runtime = 0:01:03.96
|
||||
predict_samples_per_second = 0.782
|
||||
predict_steps_per_second = 0.063
|
||||
08/18/2024 17:42:52 - INFO - llamafactory.train.sft.trainer - Saving prediction results to ./results/inference/Llama2-7B/llama2_predict_3/generated_predictions.jsonl
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Llama2-7B/llama2_predict_3
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.0031,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.9636,
|
||||
"predict_samples_per_second": 0.782,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.4552321428571429,
|
||||
"predict_model_preparation_time": 0.0031,
|
||||
"predict_rouge-1": 11.098439285714287,
|
||||
"predict_rouge-2": 3.044635714285714,
|
||||
"predict_rouge-l": 5.031203571428571,
|
||||
"predict_runtime": 63.9636,
|
||||
"predict_samples_per_second": 0.782,
|
||||
"predict_steps_per_second": 0.063
|
||||
}
|
||||
|
|
@ -1,20 +0,0 @@
|
|||
08/18/2024 17:56:23 - INFO - llamafactory.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, compute dtype: None
|
||||
08/18/2024 17:56:24 - INFO - llamafactory.data.loader - Loading dataset llamafactory/alpaca_gpt4_zh...
|
||||
eval example:
|
||||
input_ids:
|
||||
[1, 518, 25580, 29962, 29871, 30982, 31695, 31863, 31577, 30210, 30457, 30502, 31302, 30858, 30267, 518, 29914, 25580, 29962]
|
||||
inputs:
|
||||
<s> [INST] 保持健康的三个提示。 [/INST]
|
||||
08/18/2024 17:56:29 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
|
||||
08/18/2024 17:56:58 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
|
||||
08/18/2024 17:56:58 - INFO - llamafactory.model.loader - all params: 6,738,415,616
|
||||
***** predict metrics *****
|
||||
predict_bleu-4 = 1.2952
|
||||
predict_model_preparation_time = 0.003
|
||||
predict_rouge-1 = 12.8381
|
||||
predict_rouge-2 = 3.2551
|
||||
predict_rouge-l = 5.1021
|
||||
predict_runtime = 0:06:01.71
|
||||
predict_samples_per_second = 0.138
|
||||
predict_steps_per_second = 0.069
|
||||
08/18/2024 18:03:00 - INFO - llamafactory.train.sft.trainer - Saving prediction results to ./results/inference/Llama2-7B/llama2_predict_3_single/generated_predictions.jsonl
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Llama2-7B/llama2_predict_3_single
|
||||
|
|
|
|||
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.2952020000000002,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 12.838101999999997,
|
||||
"predict_rouge-2": 3.255076,
|
||||
"predict_rouge-l": 5.102102,
|
||||
"predict_runtime": 361.7122,
|
||||
"predict_samples_per_second": 0.138,
|
||||
"predict_steps_per_second": 0.069
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
|
|
@ -1,10 +0,0 @@
|
|||
{
|
||||
"predict_bleu-4": 1.2952020000000002,
|
||||
"predict_model_preparation_time": 0.003,
|
||||
"predict_rouge-1": 12.838101999999997,
|
||||
"predict_rouge-2": 3.255076,
|
||||
"predict_rouge-l": 5.102102,
|
||||
"predict_runtime": 361.7122,
|
||||
"predict_samples_per_second": 0.138,
|
||||
"predict_steps_per_second": 0.069
|
||||
}
|
||||
|
|
@ -1,5 +0,0 @@
|
|||
{"current_steps": 5, "total_steps": 25, "percentage": 20.0, "elapsed_time": "0:00:52", "remaining_time": "0:03:29"}
|
||||
{"current_steps": 10, "total_steps": 25, "percentage": 40.0, "elapsed_time": "0:02:05", "remaining_time": "0:03:07"}
|
||||
{"current_steps": 15, "total_steps": 25, "percentage": 60.0, "elapsed_time": "0:03:18", "remaining_time": "0:02:12"}
|
||||
{"current_steps": 20, "total_steps": 25, "percentage": 80.0, "elapsed_time": "0:04:31", "remaining_time": "0:01:07"}
|
||||
{"current_steps": 25, "total_steps": 25, "percentage": 100.0, "elapsed_time": "0:05:44", "remaining_time": "0:00:00"}
|
||||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Qwen-7B/Qwen_predict_1
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Qwen-7B/Qwen_predict_1_single
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Qwen-7B/Qwen_predict_2
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Qwen-7B/Qwen_predict_2_single
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Qwen-7B/Qwen_predict_3
|
||||
|
|
|
|||
|
|
@ -11,6 +11,7 @@ cutoff_len: 1024
|
|||
max_samples: 50
|
||||
overwrite_cache: true
|
||||
preprocessing_num_workers: 16
|
||||
include_tokens_per_second: true
|
||||
|
||||
### output
|
||||
output_dir: ./results/inference/Qwen-7B/Qwen_predict_3_single
|
||||
|
|
|
|||
|
|
@ -15,6 +15,7 @@
|
|||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import time
|
||||
from typing import TYPE_CHECKING, List, Optional
|
||||
|
||||
from ...data import SFTDataCollatorWith4DAttentionMask, get_dataset
|
||||
|
|
@ -112,7 +113,9 @@ def run_sft(
|
|||
|
||||
# Predict
|
||||
if training_args.do_predict:
|
||||
print("predict start time: " + time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
|
||||
predict_results = trainer.predict(dataset_module["eval_dataset"], metric_key_prefix="predict", **gen_kwargs)
|
||||
print("predict end time: " + time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
|
||||
if training_args.predict_with_generate: # predict_loss will be wrong if predict_with_generate is enabled
|
||||
predict_results.metrics.pop("predict_loss", None)
|
||||
trainer.log_metrics("predict", predict_results.metrics)
|
||||
|
|
|
|||
Loading…
Reference in New Issue