Enable the export of only one decoder #1257
ONNX export decoder model refactorization
41b8f981
fix style
f91a0184
fix index
4ce5fbe0
merge main in branch
552eebc9
Merge branch 'main' into refactorization-decoder-ort
aa40ba47
fix IO bindings
9fa05e4c
format
3a0d76ab
enable mpt support
b0aa2341
format
dfabefd7
add trust remote code
35df7bde
fix test
469edc83
format
77cc527c
rm redundant
4f72a7eb
format
599c31c1
merge main in branch
dac2376b
fix
c13b6455
Merge branch 'main' into refactorization-decoder-ort
0e83cd18
Merge branch 'main' into refactorization-decoder-ort
1f81f0b1
fix quantization
a0d0802d
add test
7f65ce1e
format
2840b81d
echarlaix
marked this pull request as ready for review 2 years ago
format
5fa7b203
fix optimization
80119828
fix opitmization
b6433086
fix compatibility with legacy models
ca9ce301
echarlaix
marked this pull request as draft 2 years ago
format
144753ac
fix legacy models
4ee61674
format
f2d0f841
fix style
3ff719a7
format
d794141a
add export to main_export
a34a16e0
add legacy to ONNX export
dfe7e5e4
fix test
8d102f78
fix
62b89742
rm unused import
b8e18c30
patch model to fix causal lm generation
819691ef
rm commen
e259670f
add no psot process
2f262019
merge main in branch
bed73d4c
fix
6d8acb42
remove bloom caching
52c17457
fix
1e9ba7e4
format
4b68caa3
fix dynamic axis for position ids
e5fd9f8e
fix external data
addad926
format
2c063c07
test
1b47093d
test
35caaf22
add model patcher
725857be
format
46b26b5f
fix
33957af8
fix bart model patcher
c2ec382a
format
d86bce64
format
be836b5d
fix model patcher for opt models
b05f5991
fix format
26d97e8f
add tmp onnxruntime max version
4b6c3ed2
add test
615a2198
format
b3525f8d
tmp fix onnxruntime max version
e0e2bae1
format
cbc935fd
add test
624d91da
fix ort docker
c5584504
fix format
e72526d4
merge main in branch
7926999a
add test
44ef0f1b
echarlaix
marked this pull request as ready for review 2 years ago
fix bart model patcher
ed8e74f1
raise when unsupported model
c13a170a
add cached file
524b6682
minor
8951ddf4
add position warning
2491ef33
fixes
0ab6e61e
enable post process after export to remove tied weights
1a7d4919
comment
cd8d4be8
remove test
e6de5e76
fix test
4a32f7a1
modify model
a51686ec
remove deprecated use_merged in test
e2f8a3b6
Merge branch 'main' into refactorization-decoder-ort
52ce2d71
Add mistral model patcher
b76f43a8
fix test
5b3d4453
add slow test
5406f95b
add workflow
52e0c699
fix
88833236
echarlaix
merged
6e157770
into main 2 years ago
echarlaix
deleted the refactorization-decoder-ort branch 2 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub