change: change yaml
This commit is contained in:
wql
parent
07b328ee23
commit
d8a730dcfe
24
batch_run.sh
24
batch_run.sh
|
|
@ -5,12 +5,12 @@ CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/Llama2-7B/llama2_
|
|||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/Llama2-7B/llama2_lora_sft_2_single.yaml | tee results/lora_sft/Llama2-7B/llama2_lora_sft_2_single.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/Llama2-7B/llama2_lora_sft_3_single.yaml | tee results/lora_sft/Llama2-7B/llama2_lora_sft_3_single.txt
|
||||
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_1.yaml | tee results/inference/Llama2-7B/llama2_predict_1.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_2.yaml | tee results/inference/Llama2-7B/llama2_predict_2.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_3.yaml | tee results/inference/Llama2-7B/llama2_predict_3.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.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.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.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_1.yaml | tee results/inference/Llama2-7B/llama2_predict_1.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_2.yaml | tee results/inference/Llama2-7B/llama2_predict_2.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/inference/Llama2-7B/llama2_predict_3.yaml | tee results/inference/Llama2-7B/llama2_predict_3.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.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.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.txt
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_1.yaml | tee results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_1.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_2.yaml | tee results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_2.txt
|
||||
|
|
@ -19,12 +19,12 @@ CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/inference/Llama2-7B/llama2
|
|||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_2_single.yaml | tee results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_2_single.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_3_single.yaml | tee results/lora_sft/Baichuan2-7B/Baichuan2_lora_sft_3_single.txt
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1_single.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1_single.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2_single.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2_single.txt
|
||||
# CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3_single.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3_single.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2.txt
|
||||
FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1_single.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_1_single.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2_single.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_2_single.txt
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3_single.yaml | tee results/lora_sft/ChatGLM2-6B/ChatGLM2_lora_sft_3_single.txt
|
||||
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/Qwen-7B/Qwen_lora_sft_1.yaml | tee results/lora_sft/Qwen-7B/Qwen_lora_sft_1.txt
|
||||
# FORCE_TORCHRUN=1 llamafactory-cli train results/lora_sft/Qwen-7B/Qwen_lora_sft_2.yaml | tee results/lora_sft/Qwen-7B/Qwen_lora_sft_2.txt
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ~/data
|
||||
|
||||
### method
|
||||
stage: sft
|
||||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ~/data
|
||||
|
||||
### method
|
||||
stage: sft
|
||||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ~/data
|
||||
|
||||
### method
|
||||
stage: sft
|
||||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ~/data
|
||||
|
||||
### method
|
||||
stage: sft
|
||||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ~/data
|
||||
|
||||
### method
|
||||
stage: sft
|
||||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
### model
|
||||
model_name_or_path: ZhipuAI/chatglm2-6b
|
||||
model_name_or_path: ~/data
|
||||
|
||||
### method
|
||||
stage: sft
|
||||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -31,7 +31,8 @@ lr_scheduler_type: cosine
|
|||
warmup_ratio: 0.1
|
||||
fp16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 500
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -31,7 +31,8 @@ lr_scheduler_type: cosine
|
|||
warmup_ratio: 0.1
|
||||
fp16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 500
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -31,7 +31,8 @@ lr_scheduler_type: cosine
|
|||
warmup_ratio: 0.1
|
||||
fp16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 500
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -31,7 +31,8 @@ lr_scheduler_type: cosine
|
|||
warmup_ratio: 0.1
|
||||
fp16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 500
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -31,7 +31,8 @@ lr_scheduler_type: cosine
|
|||
warmup_ratio: 0.1
|
||||
fp16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 500
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -31,7 +31,8 @@ lr_scheduler_type: cosine
|
|||
warmup_ratio: 0.1
|
||||
fp16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 500
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
|
|
@ -32,6 +32,7 @@ warmup_ratio: 0.1
|
|||
bf16: true
|
||||
ddp_timeout: 180000000
|
||||
max_steps: 1000
|
||||
include_num_input_tokens_seen: true
|
||||
|
||||
### eval
|
||||
val_size: 0.1
|
||||
|
|
|
|||
Loading…
Reference in New Issue