ipex-llm/python/llm/test/run-llm-inference-tests-gpu.sh
Yina Chen 98b86f83d4 Support fast rope for training (#9745)
* init

* init

* fix style

* add test and fix

* address comment

* update

* merge upstream main
2024-01-17 15:51:38 +08:00

37 lines
889 B
Bash

#!/bin/bash
export ANALYTICS_ZOO_ROOT=${ANALYTICS_ZOO_ROOT}
export LLM_HOME=${ANALYTICS_ZOO_ROOT}/python/llm/src
export LLM_INFERENCE_TEST_DIR=${ANALYTICS_ZOO_ROOT}/python/llm/test/inference_gpu
export USE_XETLA=OFF
export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
export DEVICE='xpu'
set -e
echo "# Start testing inference"
start=$(date "+%s")
if [ -z "$THREAD_NUM" ]; then
THREAD_NUM=2
fi
export OMP_NUM_THREADS=$THREAD_NUM
pytest ${LLM_INFERENCE_TEST_DIR}/test_transformers_api.py -v -s
now=$(date "+%s")
time=$((now-start))
echo "Bigdl-llm gpu inference tests finished"
echo "Time used:$time seconds"
echo "# Start testing layers.fast_rope_embedding"
start=$(date "+%s")
pytest ${LLM_INFERENCE_TEST_DIR}/test_layer_fast_rope.py -v -s
now=$(date "+%s")
time=$((now-start))
echo "Bigdl-llm gpu layers.fast_rope_embedding tests finished"
echo "Time used:$time seconds"