You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2022/11/25 15:21:19 UTC
[GitHub] [tvm] leandron opened a new pull request, #13487: [ACL] Enable int8 data type in QNN DENSE
leandron opened a new pull request, #13487:
URL: https://github.com/apache/tvm/pull/13487
This enables `QNN_DENSE` int8 data type to be used in Compute Library for the Arm(r) Architecture (ACL) BYOC integration.
cc @lhutton1 for reviews
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] tvm-bot commented on pull request #13487: [ACL] Enable int8 data type in QNN DENSE
Posted by GitBox <gi...@apache.org>.
tvm-bot commented on PR #13487:
URL: https://github.com/apache/tvm/pull/13487#issuecomment-1327623129
<!---bot-comment-->
Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews from [Reviewers](https://github.com/apache/incubator-tvm/blob/master/CONTRIBUTORS.md#reviewers) by @-ing them in a comment.
<!--bot-comment-ccs-start-->
* cc @ashutosh-arm, @lhutton1 <sub>See [#10317](https://github.com/apache/tvm/issues/10317) for details</sub><!--bot-comment-ccs-end-->
<sub>Generated by [tvm-bot](https://github.com/apache/tvm/blob/main/ci/README.md#github-actions)</sub>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #13487: [ACL] Enable int8 data type in QNN DENSE
Posted by GitBox <gi...@apache.org>.
ashutosh-arm commented on code in PR #13487:
URL: https://github.com/apache/tvm/pull/13487#discussion_r1033320858
##########
tests/python/contrib/test_arm_compute_lib/test_dense.py:
##########
@@ -251,7 +252,14 @@ def test_codegen_dense():
verify_codegen(func, exp_codegen)
-def test_qnn_dense():
+@pytest.mark.parametrize(
+ "dtype,min_range,max_range",
+ [
+ ("uint8", 0, 255),
+ ("int8", -127, 128),
+ ],
+)
Review Comment:
Yes, moving this into TVM testing area would be more beneficial.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #13487: [ACL] Enable int8 data type in QNN DENSE
Posted by GitBox <gi...@apache.org>.
ashutosh-arm commented on code in PR #13487:
URL: https://github.com/apache/tvm/pull/13487#discussion_r1032693187
##########
tests/python/contrib/test_arm_compute_lib/test_dense.py:
##########
@@ -251,7 +252,14 @@ def test_codegen_dense():
verify_codegen(func, exp_codegen)
-def test_qnn_dense():
+@pytest.mark.parametrize(
+ "dtype,min_range,max_range",
+ [
+ ("uint8", 0, 255),
+ ("int8", -127, 128),
+ ],
+)
Review Comment:
For future usage, could we define a function either with acl tests / under tvm testing that would provide the ranges?
Similar to https://github.com/apache/tvm/blob/2105b937cd6b4f426156813d6562f2845e5604f0/tests/python/contrib/test_cmsisnn/utils.py#L89
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] lhutton1 commented on a diff in pull request #13487: [ACL] Enable int8 data type in QNN DENSE
Posted by GitBox <gi...@apache.org>.
lhutton1 commented on code in PR #13487:
URL: https://github.com/apache/tvm/pull/13487#discussion_r1033297838
##########
tests/python/contrib/test_arm_compute_lib/test_dense.py:
##########
@@ -251,7 +252,14 @@ def test_codegen_dense():
verify_codegen(func, exp_codegen)
-def test_qnn_dense():
+@pytest.mark.parametrize(
+ "dtype,min_range,max_range",
+ [
+ ("uint8", 0, 255),
+ ("int8", -127, 128),
+ ],
+)
Review Comment:
+1, I intended to do this for NPU testing as well but never got around to it, perhaps we can pull it out into a bigger refactoring?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] leandron commented on a diff in pull request #13487: [ACL] Enable int8 data type in QNN DENSE
Posted by GitBox <gi...@apache.org>.
leandron commented on code in PR #13487:
URL: https://github.com/apache/tvm/pull/13487#discussion_r1033227736
##########
tests/python/contrib/test_arm_compute_lib/test_dense.py:
##########
@@ -251,7 +252,14 @@ def test_codegen_dense():
verify_codegen(func, exp_codegen)
-def test_qnn_dense():
+@pytest.mark.parametrize(
+ "dtype,min_range,max_range",
+ [
+ ("uint8", 0, 255),
+ ("int8", -127, 128),
+ ],
+)
Review Comment:
That's a good idea, I can start moving that direction when implementing the next operators.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] lhutton1 merged pull request #13487: [ACL] Enable int8 data type in QNN DENSE
Posted by GitBox <gi...@apache.org>.
lhutton1 merged PR #13487:
URL: https://github.com/apache/tvm/pull/13487
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org