update bash

TongLi3701 · Jul 12, 2024 · 94ddbf3 · 94ddbf3
1 parent bd861a1
commit 94ddbf3
Show file tree

Hide file tree

Showing 5 changed files with 5 additions and 13 deletions.
diff --git a/applications/ColossalChat/examples/training_scripts/train_dpo.sh b/applications/ColossalChat/examples/training_scripts/train_dpo.sh
@@ -15,9 +15,8 @@ set_n_least_used_CUDA_VISIBLE_DEVICES() {
 }
 set_n_least_used_CUDA_VISIBLE_DEVICES 4
 
-PROJECT_NAME="dpo"
+PROJECT_NAME="DPO"
 PARENT_SAVE_DIR="" # Path to a folder to save checkpoints
-PARENT_TENSORBOARD_DIR="" # Path to a folder to save logs
 PARENT_CONFIG_FILE="" # Path to a folder to save training config logs
 PRETRAINED_MODEL_PATH="" # huggingface or local model path
 PRETRAINED_TOKENIZER_PATH="" # huggingface or local tokenizer path
@@ -42,7 +41,6 @@ CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"
 
 colossalai run --nproc_per_node 4 --hostfile hostfile --master_port 31313 train_dpo.py \
     --pretrain $PRETRAINED_MODEL_PATH \
-    --checkpoint_path $PRETRAINED_MODEL_PATH \
     --tokenizer_dir $PRETRAINED_TOKENIZER_PATH \
     --dataset ${dataset[@]} \
     --plugin "zero2" \

diff --git a/applications/ColossalChat/examples/training_scripts/train_orpo.sh b/applications/ColossalChat/examples/training_scripts/train_orpo.sh
@@ -15,9 +15,8 @@ set_n_least_used_CUDA_VISIBLE_DEVICES() {
 }
 set_n_least_used_CUDA_VISIBLE_DEVICES 2
 
-PROJECT_NAME="dpo"
+PROJECT_NAME="ORPO"
 PARENT_SAVE_DIR="" # Path to a folder to save checkpoints
-PARENT_TENSORBOARD_DIR="" # Path to a folder to save logs
 PARENT_CONFIG_FILE="" # Path to a folder to save training config logs
 PRETRAINED_MODEL_PATH="" # huggingface or local model path
 PRETRAINED_TOKENIZER_PATH="" # huggingface or local tokenizer path
@@ -42,7 +41,6 @@ CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"
 
 colossalai run --nproc_per_node 8 --hostfile hostfile --master_port 31313 train_orpo.py \
     --pretrain $PRETRAINED_MODEL_PATH \
-    --checkpoint_path $PRETRAINED_MODEL_PATH \
     --tokenizer_dir $PRETRAINED_TOKENIZER_PATH \
     --dataset ${dataset[@]} \
     --plugin "zero2" \

diff --git a/applications/ColossalChat/examples/training_scripts/train_ppo.sh b/applications/ColossalChat/examples/training_scripts/train_ppo.sh
@@ -15,10 +15,9 @@ set_n_least_used_CUDA_VISIBLE_DEVICES() {
 }
 set_n_least_used_CUDA_VISIBLE_DEVICES 8
 
-PROJECT_NAME="ppo"
+PROJECT_NAME="PPO"
 
 PARENT_SAVE_DIR="" # Path to a folder to save checkpoints
-PARENT_TENSORBOARD_DIR="" # Path to a folder to save logs
 PARENT_CONFIG_FILE="" # Path to a folder to save training config logs
 PRETRAINED_MODEL_PATH="" # local pretrained model path (from RLHF step 1: SFT)
 PRETRAINED_TOKENIZER_PATH="" # huggingface or local tokenizer path

diff --git a/applications/ColossalChat/examples/training_scripts/train_rm.sh b/applications/ColossalChat/examples/training_scripts/train_rm.sh
@@ -15,9 +15,8 @@ set_n_least_used_CUDA_VISIBLE_DEVICES() {
 }
 set_n_least_used_CUDA_VISIBLE_DEVICES 8
 
-PROJECT_NAME="rm"
+PROJECT_NAME="RM"
 PARENT_SAVE_DIR="" # Path to a folder to save checkpoints
-PARENT_TENSORBOARD_DIR="" # Path to a folder to save logs
 PARENT_CONFIG_FILE="" # Path to a folder to save training config logs
 PRETRAINED_MODEL_PATH="" # huggingface or local model path
 PRETRAINED_TOKENIZER_PATH="" # huggingface or local tokenizer path
@@ -42,7 +41,6 @@ CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"
 
 colossalai run --nproc_per_node 8 --hostfile hostfile --master_port 31312 train_rm.py \
     --pretrain $PRETRAINED_MODEL_PATH \
-    --checkpoint_path /home/yeanbang/data/experiments/rm/hhh_aligh/ckptllama2-rm-2024-01-17-14-43-24/epoch-1_step-1317/modeling \
     --tokenizer_dir $PRETRAINED_TOKENIZER_PATH \
     --dataset ${dataset[@]} \
     --plugin "zero2" \

diff --git a/applications/ColossalChat/examples/training_scripts/train_sft.sh b/applications/ColossalChat/examples/training_scripts/train_sft.sh
@@ -14,9 +14,8 @@ set_n_least_used_CUDA_VISIBLE_DEVICES() {
 }
 
 set_n_least_used_CUDA_VISIBLE_DEVICES 4
-PROJECT_NAME="sft"
+PROJECT_NAME="SFT"
 PARENT_SAVE_DIR="" # Path to a folder to save checkpoints
-PARENT_TENSORBOARD_DIR="" # Path to a folder to save logs
 PARENT_CONFIG_FILE="" # Path to a folder to save training config logs
 PRETRAINED_MODEL_PATH="" # huggingface or local model path
 PRETRAINED_TOKENIZER_PATH="" # huggingface or local tokenizer path