[TOPI][Hexagon] Implement quantized adaptive_avg_pool1d for hexagon #13282
Conversation
|
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 by @-ing them in a comment. Generated by tvm-bot |
trahman-quic
changed the title
|
@trahman-quic pylint has been added to the test files. Please fix the lint issues |
@mehrdadh The checks are green now. Wanted to ask you if it's good to merge now |
| # KIND, either express or implied. See the License for the | ||
| # specific language governing permissions and limitations | ||
| # under the License. | ||
| # pylint: disable=invalid-name, unused-variable, unused-argument, too-many-locals |
There was a problem hiding this comment.
I don't think we need these. Please remove them and fix the lint issues.
| @@ -0,0 +1,181 @@ | |||
| # Licensed to the Apache Software Foundation (ASF) under one | |||
There was a problem hiding this comment.
I have added lint for tests in this sub-directory. Let's wait for that PR to merge to avoid conflict. After that merges, you will see lint issues with this test, so I recommend to fix it in the meantime
#13271
trahman-quic
force-pushed
the
trahman/adaptive_avg_pool1d
branch
from
5ef9168 to
b93d61b
Compare
|
Hi @mehrdadh The lint issues have been fixed but there are some CI tests that are failing. Are these related to my tests? Can you please give me an idea on how to resolve the issues? |
|
@trahman-quic |
|
Hi @mehrdadh thank you for suggesting the fix for import. All tests have passed now. If it looks good to you, we can merge it. |
…pache#13282) * [TOPI][Hexagon] Implement adaptive_avg_pool1d for hexagon * Fix lint issues * Fix some lint issues * Fix lint issues in test * Fix import for allocate_hexagon_array
4 participants
This PR adds the implementation, test cases and schedules of quantized adaptive_avg_pool1d for hexagon. Input is assumed to be in NCW layout with a size of multiple of croutons . Here we are introducing a new crouton layout ncw-32c64w-2d. The op implementation is for specialized case where output_size is NxCx1