a644350589
Tacotron2+Waveglow/PyT * AMP support * Data preprocessing for Tacotron 2 training * Fixed dropouts on LSTMCells SSD/PyT * script and notebook for inference * AMP support * README update * updates to examples/* BERT/PyT * initial release GNMT/PyT * Default container updated to NGC PyTorch 19.05-py3 * Mixed precision training implemented using APEX AMP * Added inference throughput and latency results on NVIDIA Tesla V100 16G * Added option to run inference on user-provided raw input text from command line NCF/PyT * Updated performance tables. * Default container changed to PyTorch 19.06-py3. * Caching validation negatives between runs Transformer/PyT * new README * jit support added UNet Medical/TF * inference example scripts added * inference benchmark measuring latency added * TRT/TF-TRT support added * README updated GNMT/TF * Performance improvements Small updates (mostly README) for other models.
45 lines
1.1 KiB
Bash
45 lines
1.1 KiB
Bash
#!/bin/bash
|
|
|
|
set -e
|
|
|
|
DATASET_DIR='data/wmt16_de_en'
|
|
REPO_DIR='/workspace/gnmt'
|
|
REFERENCE_FILE=$REPO_DIR/scripts/tests/reference_inference_performance
|
|
|
|
MATH=$1
|
|
if [[ ${MATH} != "fp16" && ${MATH} != "fp32" ]]; then
|
|
echo "Unsupported option for MATH, use either 'fp16' or 'fp32'"
|
|
exit 1
|
|
fi
|
|
|
|
BATCH_SIZE=128
|
|
BEAM_SIZE=5
|
|
PERF_TOLERANCE=0.95
|
|
|
|
GPU_NAME=`nvidia-smi --query-gpu=gpu_name --format=csv,noheader |uniq`
|
|
echo 'GPU_NAME:' ${GPU_NAME}
|
|
|
|
REFERENCE_PERF=`grep "${MATH},${BATCH_SIZE},${BEAM_SIZE},${GPU_NAME}" \
|
|
${REFERENCE_FILE} | \cut -f 5 -d ','`
|
|
|
|
if [ -z "${REFERENCE_PERF}" ]; then
|
|
echo "WARNING: COULD NOT FIND REFERENCE PERFORMANCE FOR EXECUTED CONFIG"
|
|
TARGET_PERF=''
|
|
else
|
|
PERF_THRESHOLD=$(awk 'BEGIN {print ('${REFERENCE_PERF}' * '${PERF_TOLERANCE}')}')
|
|
TARGET_PERF='--target-perf '${PERF_THRESHOLD}
|
|
fi
|
|
|
|
cd $REPO_DIR
|
|
|
|
python3 translate.py \
|
|
--input ${DATASET_DIR}/newstest2014.en \
|
|
--reference ${DATASET_DIR}/newstest2014.de \
|
|
--output /tmp/output \
|
|
--model results/gnmt/model_best.pth \
|
|
--batch-size ${BATCH_SIZE} \
|
|
--beam-size ${BEAM_SIZE} \
|
|
--math ${MATH} \
|
|
--warmup 1 \
|
|
--target-bleu 24.3 \
|
|
${TARGET_PERF}
|