ros/LLaMA-Factory
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update trainers

Browse Source 
Former-commit-id: d0dd6eefed0b86895ed00a7cafb331e5193db645
This commit is contained in:
hiyouga
2024-03-28 18:16:27 +08:00
parent dc540dfaa8
commit 59e6ebf039
13 changed files with 89 additions and 145 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
examples/extras/galore/adamw.sh
View File

@@ -1,31 +0,0 @@
#!/bin/bash
CUDA_VISIBLE_DEVICES=0 python ../../../src/train_bash.py \
--stage sft \
--do_train \
--model_name_or_path meta-llama/Llama-2-7b-hf \
--dataset alpaca_gpt4_en,glaive_toolcall \
--dataset_dir ../../../data \
--template default \
--finetuning_type full \
--output_dir ../../../saves/LLaMA2-7B/galore/sft \
--overwrite_cache \
--overwrite_output_dir \
--cutoff_len 1024 \
--preprocessing_num_workers 16 \
--per_device_train_batch_size 1 \
--per_device_eval_batch_size 1 \
--gradient_accumulation_steps 1 \
--lr_scheduler_type cosine \
--logging_steps 10 \
--warmup_steps 20 \
--save_steps 100 \
--eval_steps 100 \
--evaluation_strategy steps \
--load_best_model_at_end \
--learning_rate 5e-5 \
--num_train_epochs 3.0 \
--max_samples 3000 \
--val_size 0.1 \
--plot_loss \
--fp16

32
examples/extras/galore/adamw_8bit_bf16.sh
View File

@@ -1,32 +0,0 @@
#!/bin/bash
CUDA_VISIBLE_DEVICES=0 python ../../../src/train_bash.py \
--stage sft \
--do_train \
--model_name_or_path meta-llama/Llama-2-7b-hf \
--dataset alpaca_gpt4_en,glaive_toolcall \
--dataset_dir ../../../data \
--template default \
--finetuning_type full \
--optim adamw_8bit \
--output_dir ../../../saves/LLaMA2-7B/galore/sft \
--overwrite_cache \
--overwrite_output_dir \
--cutoff_len 1024 \
--preprocessing_num_workers 16 \
--per_device_train_batch_size 1 \
--per_device_eval_batch_size 1 \
--gradient_accumulation_steps 1 \
--lr_scheduler_type cosine \
--logging_steps 10 \
--warmup_steps 20 \
--save_steps 100 \
--eval_steps 100 \
--evaluation_strategy steps \
--load_best_model_at_end \
--learning_rate 5e-5 \
--num_train_epochs 3.0 \
--max_samples 3000 \
--val_size 0.1 \
--plot_loss \
--pure_bf16

36
examples/extras/galore/galore_adamw_8bit_bf16.sh
View File

@@ -1,36 +0,0 @@
#!/bin/bash
CUDA_VISIBLE_DEVICES=0 python ../../../src/train_bash.py \
--stage sft \
--do_train \
--model_name_or_path meta-llama/Llama-2-7b-hf \
--dataset alpaca_gpt4_en,glaive_toolcall \
--dataset_dir ../../../data \
--template default \
--finetuning_type full \
--optim adamw_8bit \
--use_galore \
--galore_layerwise \
--galore_target mlp,self_attn \
--galore_rank 128 \
--output_dir ../../../saves/LLaMA2-7B/galore/sft \
--overwrite_cache \
--overwrite_output_dir \
--cutoff_len 1024 \
--preprocessing_num_workers 16 \
--per_device_train_batch_size 1 \
--per_device_eval_batch_size 1 \
--gradient_accumulation_steps 1 \
--lr_scheduler_type cosine \
--logging_steps 10 \
--warmup_steps 20 \
--save_steps 100 \
--eval_steps 100 \
--evaluation_strategy steps \
--load_best_model_at_end \
--learning_rate 5e-5 \
--num_train_epochs 3.0 \
--max_samples 3000 \
--val_size 0.1 \
--plot_loss \
--pure_bf16

2
examples/extras/galore/galore_adamw.sh → examples/extras/galore/sft.sh
View File

@@ -32,4 +32,4 @@ CUDA_VISIBLE_DEVICES=0 python ../../../src/train_bash.py \
--max_samples 3000 \
--val_size 0.1 \
--plot_loss \
--fp16
--pure_bf16