Skip to content

Commit

Permalink
fix issue
Browse files Browse the repository at this point in the history
  • Loading branch information
TongLi3701 committed Jul 12, 2024
1 parent 14bd4d2 commit bd861a1
Show file tree
Hide file tree
Showing 5 changed files with 6 additions and 6 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ declare -a dataset=(
TIMESTAMP=$(date +%Y-%m-%d-%H-%M-%S)
FULL_PROJECT_NAME="${PROJECT_NAME}-${TIMESTAMP}"
SAVE_DIR="${PARENT_SAVE_DIR}${FULL_PROJECT_NAME}"
CONFIG_FILE="${PARENT_CONFIG_FILE}-${FULL_PROJECT_NAME}.json"
CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"

colossalai run --nproc_per_node 4 --hostfile hostfile --master_port 31313 train_dpo.py \
--pretrain $PRETRAINED_MODEL_PATH \
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ set_n_least_used_CUDA_VISIBLE_DEVICES() {
echo "Now CUDA_VISIBLE_DEVICES is set to:"
echo "CUDA_VISIBLE_DEVICES=$CUDA_VISIBLE_DEVICES"
}
set_n_least_used_CUDA_VISIBLE_DEVICES 8
set_n_least_used_CUDA_VISIBLE_DEVICES 2

PROJECT_NAME="dpo"
PARENT_SAVE_DIR="" # Path to a folder to save checkpoints
Expand All @@ -38,7 +38,7 @@ declare -a dataset=(
TIMESTAMP=$(date +%Y-%m-%d-%H-%M-%S)
FULL_PROJECT_NAME="${PROJECT_NAME}-${TIMESTAMP}"
SAVE_DIR="${PARENT_SAVE_DIR}${FULL_PROJECT_NAME}"
CONFIG_FILE="${PARENT_CONFIG_FILE}-${FULL_PROJECT_NAME}.json"
CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"

colossalai run --nproc_per_node 8 --hostfile hostfile --master_port 31313 train_orpo.py \
--pretrain $PRETRAINED_MODEL_PATH \
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,7 @@ declare -a ptx_dataset=(
TIMESTAMP=$(date +%Y-%m-%d-%H-%M-%S)
FULL_PROJECT_NAME="${PROJECT_NAME}-${TIMESTAMP}"
SAVE_DIR="${PARENT_SAVE_DIR}${FULL_PROJECT_NAME}"
CONFIG_FILE="${PARENT_CONFIG_FILE}-${FULL_PROJECT_NAME}.json"
CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"

colossalai run --nproc_per_node 8 --hostfile hostfile --master_port 31312 train_ppo.py \
--pretrain $PRETRAINED_MODEL_PATH \
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ declare -a dataset=(
TIMESTAMP=$(date +%Y-%m-%d-%H-%M-%S)
FULL_PROJECT_NAME="${PROJECT_NAME}-${TIMESTAMP}"
SAVE_DIR="${PARENT_SAVE_DIR}${FULL_PROJECT_NAME}"
CONFIG_FILE="${PARENT_CONFIG_FILE}-${FULL_PROJECT_NAME}.json"
CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"

colossalai run --nproc_per_node 8 --hostfile hostfile --master_port 31312 train_rm.py \
--pretrain $PRETRAINED_MODEL_PATH \
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ declare -a dataset=(
TIMESTAMP=$(date +%Y-%m-%d-%H-%M-%S)
FULL_PROJECT_NAME="${PROJECT_NAME}-${TIMESTAMP}"
SAVE_DIR="${PARENT_SAVE_DIR}${FULL_PROJECT_NAME}"
CONFIG_FILE="${PARENT_CONFIG_FILE}-${FULL_PROJECT_NAME}.json"
CONFIG_FILE="${PARENT_CONFIG_FILE}${FULL_PROJECT_NAME}.json"

echo $(which colossalai)
echo $(which python)
Expand Down

0 comments on commit bd861a1

Please sign in to comment.