dfb00e37e9
* add correctness test on arc for llama model * modify layer name * add falcon ut * refactor and add ut for falcon model * modify lambda positions and update docs * replace loading pre input with last decodelayer output * switch lower bound to single model instead of using the common one * make the code implementation simple * fix gpu action allocation memory issue
25 lines
530 B
Bash
25 lines
530 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=${ANALYTICS_ZOO_ROOT}/python/llm/test/inference_gpu
|
|
|
|
export USE_XETLA=OFF
|
|
export DEVICE='xpu'
|
|
|
|
set -e
|
|
|
|
echo "# Start testing inference"
|
|
start=$(date "+%s")
|
|
|
|
if [ -z "$THREAD_NUM" ]; then
|
|
THREAD_NUM=2
|
|
fi
|
|
export OMP_NUM_THREADS=$THREAD_NUM
|
|
pytest ${LLM_INFERENCE_TEST_DIR} -v -s
|
|
|
|
now=$(date "+%s")
|
|
time=$((now-start))
|
|
|
|
echo "Bigdl-llm gpu tests finished"
|
|
echo "Time used:$time seconds"
|