[TensorRT EP] Customizable engine cache prefix #19083
yf711
merged
79 commits into
main
from
yifanl/trtep_engine_name
update
f8099b17
fix bug
ee81de54
update
452a6295
remove redundant check
6f18d8db
remove unused variable
94307680
add script to generate epcontext node
e9507e50
fix bug
35a4d333
update
44a7cc55
update
b2fdb06c
refactor
eeb65522
update
1b5117da
update
df7ef469
change function name
993b2ad5
Merge branch 'main' into chi/trt_engine_wrapper
34a86d71
update
9631f737
update
93f9fbba
refactor
d5974fc8
check compute capability
7202b73c
Merge branch 'main' into chi/trt_engine_wrapper
60f6e7ee
update for reading engine byte data
5838143e
add script to generate engine wrapper onnx model
aea26b12
refactor
f4b38f7d
fix bug
de9f5107
fix format
65331edc
refactor
befba024
add unit test
5103ddab
refactor script
f933379b
fix format
11fd2120
update
561b0599
fix format
789efe86
fix format
a5843c29
fix gen_trt_engine_wrapper_onnx_model.py
11ce3fc2
refactor unit test
f9206f38
update
0bd5b8ce
Merge branch 'main' into chi/trt_engine_wrapper
e2c3f165
update script
6605fd46
fix format
699d5381
fix bug for conflict resolve
1cbe3e9a
refactor script
f8e775d8
Merge branch 'yifanl/trt_engine_cache_prefix'
3a282a29
test
5fed04b6
fix
d1255eb0
test
5882d746
generate ep context node model from TRT EP
8f7c7ac5
add trt_dump_ep_context_model, trt_ep_context_embed_mode, trt_ep_cont…
77a62f25
fix format
04baad7e
swap the position of CreateNodeComputeFromGraph and CreateNodeCompute…
c3a028b4
merge PR 18879 and PR 18834
db46b64b
merge PR 18879 and PR 18834 (continue)
842cdf00
Merge branch 'main' into chi/trt_engine_wrapper
ca8d49fc
remove kernelcontext_setoutput
e0d33461
fix bugs after merge main
f9231a5d
Merge branch 'main' into yifanl/trtep_engine_name
389f7bb9
apply lintrunner -a
cebfcd82
cache suffix
852ccb64
fix
4a2b46f7
update
5b65e620
test
46e0a689
update
635f39a1
update
deee0fba
fix
e89d47b9
fix
883a3cb1
detail
be06261f
clean
10ec7237
add usage
a623a2d3
Use 'hardware_architecture'
b1c43055
merge main and also enforce get compute capability once inside ep con…
04359717
Merge branch 'main' into chi/trt_engine_wrapper
77bf0771
remove unnecessary code
28bdd0a0
Merge branch 'chi/trt_engine_wrapper' into yifanl/trt_engine_wrapper+…
ef9103e8
Merge branch 'chi/trt_engine_wrapper' into yifanl/trt_engine_wrapper+…
9a73eb92
sync
d00173a9
sync
ef324093
Merge branch 'main' into yifanl/trt_engine_wrapper+customized_engine_…
440e87fb
unit test
0f5fb5c1
lint
e7ccc9f9
lint
21481296
fix
8224b18f
chilo-ms
commented
on 2024-01-12
chilo-ms
commented
on 2024-01-12
apply comments
6ebb1a31
chilo-ms
commented
on 2024-01-12
chilo-ms
approved these changes
on 2024-01-12
jywu-msft
approved these changes
on 2024-01-13
Login to write a write a comment.
Login via GitHub