onnxruntime
Quantization tool: support float 8 with MatMul, support float 16 weights
#18043
Merged
Overview
Commits
47
Changes
View On GitHub

Quantization tool: support float 8 with MatMul, support float 16 weights #18043

xadupre merged 47 commits into microsoft:main from xadupre:qdqmm
xadupre
xadupre Update quantization tools to support MatMul with float 8
f09ad3cb
xadupre support float16
bd47221e
github-advanced-security
github-advanced-security commented on 2023-10-20
xadupre more consistent with types
23c9e393
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
7e7d0fe2
xadupre fix types
c073b88b
xadupre fix many unit tests
35db1499
github-advanced-security
github-advanced-security commented on 2023-10-24
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
799f7c54
xadupre fix conversion, rounding
3a14a313
xadupre new fixes
8199196a
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
87e2ba17
xadupre fix softmax qdq
6003524f
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
aa11d25b
xadupre fix shape info
0e966688
xadupre update test
170e5c9b
xadupre fix remaining unit tests
47b41b60
xadupre add value_info
b15ffcdd
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
01636994
xadupre add subtest
34b7a388
xadupre refactoring onnxruntime/test/python/quantization/test_op_matmul.py
8711792d
xadupre disable f16 for old onnx package
4806a041
xadupre disable f16 unit tests
260dd59e
xadupre support for Conv and float 16
d376f664
xadupre extend unit test for Conv
76d92843
xadupre fix lint
d2f9294b
github-advanced-security
github-advanced-security commented on 2023-10-27
xadupre xadupre marked this pull request as ready for review 2 years ago
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
6702b81f
xadupre change the disable condition
a6433e88
xadupre lint
301c435f
xadupre final fix
1e70b3e0
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
9049bcd2
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
e4f0415b
xadupre fir merge conflicts
47efb3e0
github-advanced-security
github-advanced-security commented on 2023-11-20
xadupre fix merge conflict
fcf40b40
github-advanced-security
github-advanced-security commented on 2023-11-20
xadupre fix missing min_real_range
bf44eba0
xadupre xadupre requested a review from yufenglee yufenglee 2 years ago
xadupre merge conflicts
81528cc7
xadupre fix constant
63b84d19
github-advanced-security
github-advanced-security commented on 2023-12-21
github-advanced-security
github-advanced-security commented on 2023-12-21
xadupre fix missing dtype
a90f9f5f
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
0164a38d
xadupre use np arrays
e6c39f4c
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
1c8ae860
xadupre improve robustness
1185de09
xadupre fix type issue
aee75fff
xadupre fix wrong types
63a8ea90
xadupre fix one bug
1948c4ad
xadupre fix dtype issue
67bab544
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
fe1d0fe0
yufenglee
yufenglee commented on 2024-01-09
yufenglee
yufenglee commented on 2024-01-09
xadupre better error message
1d49bc51
xadupre Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
fc406f9d
yufenglee
yufenglee approved these changes on 2024-01-11
xadupre xadupre merged c8399a81 into main 2 years ago
xadupre xadupre deleted the qdqmm branch 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum