root commited on
Commit ·
058a116
1
Parent(s): 012fc62
add files
Browse files- encoder-epoch-99-avg-1.onnx +3 -0
- export_trt_9_0.log +0 -0
- export_trt_9_0_0_2.sh +27 -0
encoder-epoch-99-avg-1.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd873c9627477e623038eabbcbe8c11aee1163157424d3925c3d6e61957db8fc
|
| 3 |
+
size 270007534
|
export_trt_9_0.log
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
export_trt_9_0_0_2.sh
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# nvcr.io/ea-bignlp/nemofw-training:23.05-py3
|
| 2 |
+
|
| 3 |
+
export CUDA_VISIBLE_DEVICES="2" # A10
|
| 4 |
+
trtexec_path=../TensorRT-9.0.0.2/bin/trtexec
|
| 5 |
+
LD_LIBRARY_PATH=/tmpdisk/yuekaiz/quant/TensorRT-9.0.0.2/lib
|
| 6 |
+
|
| 7 |
+
onnx_model=./encoder-epoch-99-avg-1.onnx
|
| 8 |
+
trt_model=./encoder.trt90.plan
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
MIN_BATCH=1
|
| 12 |
+
OPT_BATCH=4
|
| 13 |
+
MAX_BATCH=8
|
| 14 |
+
|
| 15 |
+
ENC_MIN_LEN=16
|
| 16 |
+
ENC_OPT_LEN=512
|
| 17 |
+
ENC_MAX_LEN=2000
|
| 18 |
+
|
| 19 |
+
trtexec \
|
| 20 |
+
--onnx=$onnx_model \
|
| 21 |
+
--minShapes=x:${MIN_BATCH}x${ENC_MIN_LEN}x80,x_lens:${MIN_BATCH} \
|
| 22 |
+
--optShapes=x:${OPT_BATCH}x${ENC_OPT_LEN}x80,x_lens:${OPT_BATCH} \
|
| 23 |
+
--maxShapes=x:${MAX_BATCH}x${ENC_MAX_LEN}x80,x_lens:${MAX_BATCH} \
|
| 24 |
+
--fp16 \
|
| 25 |
+
--skipInference \
|
| 26 |
+
--verbose \
|
| 27 |
+
--saveEngine=$trt_model > export_trt_9_0.log
|