github
/
ColossalAI
mirror of https://github.com/hpcaitech/ColossalAI
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

[example] fix benchmark.sh for gpt example (#2229)

pull/2234/head
HELSON 2022-12-29 23:00:14 +08:00 committed by GitHub
parent 78483a9fdd
commit 31fe84237b
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
examples/language/gpt/benchmark.sh
View File

@ -1,6 +1,6 @@
for MODEL_NAME in "GPT2small"
for MODEL_TYPE in "gpt2_medium"
do
for BATCH_SIZE in 8
for BATCH_SIZE in 16
do
for GPUNUM in 1 2 4 8
do
@ -11,8 +11,8 @@ then
continue
fi
echo "****************** Begin ***************************"
echo "* benchmrking MODEL_NAME ${MODEL_NAME} BS ${BATCH_SIZE} BS ${BS} GPUNUM ${GPUNUM} TPDEGREE ${TPDEGREE}"
bash ./run.sh
echo "* benchmrking MODEL_TYPE ${MODEL_TYPE} BS ${BATCH_SIZE} BS ${BS} GPUNUM ${GPUNUM} TPDEGREE ${TPDEGREE}"
MODEL_TYPE=${MODEL_TYPE} BATCH_SIZE=${BATCH_SIZE} GPUNUM=${GPUNUM} TPDEGREE=${TPDEGREE} bash ./run.sh
echo "****************** Finished ***************************"
echo ""
echo ""

12
examples/language/gpt/run.sh
View File

@ -1,13 +1,13 @@
# distplan in ["colossalai", "zero1", "zero2", "torch_ddp", "torch_zero"]
export DISTPAN={$DISTPAN:-"colossalai"}
export DISTPAN=${DISTPAN:-"colossalai"}
# The following options only valid when DISTPAN="colossalai"
export TPDEGREE=${TPDEGREE:-1}
export GPUNUM=${GPUNUM:-1}
export PLACEMENT=${PLACEMENT:'const'}
export USE_SHARD_INIT=${USE_SHARD_INIT:False}
export BATCH_SIZE=${BATCH_SIZE:-8}
export MODEL_TYPE=${MODEL_TYPE:"gpt2_medium"}
export TPDEGREE=${TPDEGREE:-1}
export PLACEMENT=${PLACEMENT:-"const"}
export USE_SHARD_INIT=${USE_SHARD_INIT:-False}
export BATCH_SIZE=${BATCH_SIZE:-16}
export MODEL_TYPE=${MODEL_TYPE:-"gpt2_medium"}
mkdir -p logs
torchrun --standalone --nproc_per_node=${GPUNUM} train_gpt_demo.py --tp_degree=${TPDEGREE} --model_type=${MODEL_TYPE} --batch_size=${BATCH_SIZE} --placement ${PLACEMENT} --shardinit ${USE_SHARD_INIT} --distplan ${DISTPAN} 2>&1 | tee ./logs/${MODEL_TYPE}_${DISTPAN}_gpu_${GPUNUM}_bs_${BATCH_SIZE}_tp_${TPDEGREE}.log