Quantization: option to disable the calculation and use of the zero point. #2938
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Alexandre Eichenberger <[email protected]>
| Value sub = create.math.sub(xFloat, zeroPointFloat); | ||
|
|
||
| Value sub; | ||
| if (!disableQuantZeroPoint && !isNoneValue(zeroPointInt)) { |
There was a problem hiding this comment.
Note: zero point is optional in dequantization, so added support for its value being none.
Signed-off-by: Alexandre Eichenberger <[email protected]>
|
Jenkins Linux amd64 Build #15643 [push] added support for no-zer... started at 09:05 |
|
Jenkins Linux s390x Build #15646 [push] added support for no-zer... started at 10:05 |
|
Jenkins Linux ppc64le Build #14673 [push] added support for no-zer... started at 10:16 |
|
Jenkins Linux amd64 Build #15643 [push] added support for no-zer... passed after 1 hr 14 min |
|
Jenkins Linux s390x Build #15646 [push] added support for no-zer... passed after 1 hr 46 min |
|
Jenkins Linux ppc64le Build #14673 [push] added support for no-zer... passed after 2 hr 13 min |
Sunny-Anand
pushed a commit
to Sunny-Anand/onnx-mlir
that referenced
this pull request
Sep 17, 2024
Signed-off-by: Alexandre Eichenberger <[email protected]> Co-authored-by: Tung D. Le <[email protected]> Signed-off-by: Sunny-Anand <[email protected]>
Sunny-Anand
added a commit
that referenced
this pull request
Sep 17, 2024
* Change lowering of onnx.IF to Krnl (#2932) * implementation Signed-off-by: chentong319 <[email protected]> * test case change Signed-off-by: chentong319 <[email protected]> * format Signed-off-by: chentong319 <[email protected]> * add test for If back Signed-off-by: chentong319 <[email protected]> * format Signed-off-by: chentong319 <[email protected]> --------- Signed-off-by: chentong319 <[email protected]> Co-authored-by: Tung D. Le <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * Update c style cast to c++ style cast (#2934) Signed-off-by: Mike Essenmacher <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * Change c style cast to c++ style cast (#2936) Signed-off-by: Mike Essenmacher <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * Add coding practices for onnx-mlir (#2935) Signed-off-by: Mike Essenmacher <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * try to use new buffer deallocation (#2919) * implementation Signed-off-by: Chen Tong <[email protected]> * comments Signed-off-by: Chen Tong <[email protected]> * format Signed-off-by: Chen Tong <[email protected]> --------- Signed-off-by: Chen Tong <[email protected]> Co-authored-by: Tung D. Le <[email protected]> Co-authored-by: Alexandre Eichenberger <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * fix requirements.txt link Signed-off-by: Sunny-Anand <[email protected]> * Reuse input buffer in lowering to krnl (#2939) * first step Signed-off-by: chentong319 <[email protected]> * cpu Signed-off-by: chentong319 <[email protected]> * options Signed-off-by: chentong319 <[email protected]> * unify Signed-off-by: chentong319 <[email protected]> * simd Signed-off-by: chentong319 <[email protected]> * comments Signed-off-by: chentong319 <[email protected]> * lit test Signed-off-by: chentong319 <[email protected]> * fix test Signed-off-by: chentong319 <[email protected]> * format Signed-off-by: chentong319 <[email protected]> * response Signed-off-by: chentong319 <[email protected]> --------- Signed-off-by: chentong319 <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * Fix GroupNorm to support Opset21 (#2928) * Group norm for opset 21 * Testing phase * Fix GroupNorm to support Opset21 --------- Signed-off-by: hamptonm1 <[email protected]> Co-authored-by: Megan Hampton <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * Update Ops documentation for ONNX 1.16.2 (#2942) * Update Ops documentation for ONNX 1.16.2 * Fix format --------- Co-authored-by: Megan Hampton <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * LLVM/StableHLO Upgrade eaa95a1 (#2943) Co-authored-by: Megan Hampton <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * added support for no-zero-point quantization (#2938) Signed-off-by: Alexandre Eichenberger <[email protected]> Co-authored-by: Tung D. Le <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> * update with main Signed-off-by: Sunny-Anand <[email protected]> --------- Signed-off-by: chentong319 <[email protected]> Signed-off-by: Sunny-Anand <[email protected]> Signed-off-by: Mike Essenmacher <[email protected]> Signed-off-by: Chen Tong <[email protected]> Signed-off-by: hamptonm1 <[email protected]> Signed-off-by: Alexandre Eichenberger <[email protected]> Signed-off-by: Sunny Anand <[email protected]> Co-authored-by: Tong Chen <[email protected]> Co-authored-by: Tung D. Le <[email protected]> Co-authored-by: Mike Essenmacher <[email protected]> Co-authored-by: Alexandre Eichenberger <[email protected]> Co-authored-by: hamptonm1 <[email protected]> Co-authored-by: Megan Hampton <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Use
--disable-quantization-zero-pointto disable the computation of the zero point in dynamic linear quantization and ignore the zero point in linear quantization and dequantization.