Add torch.log10 INT support to ExecuTorch Arm backend (#18671)#18671
Add torch.log10 INT support to ExecuTorch Arm backend (#18671)#18671JorickvdHoeven wants to merge 1 commit intomainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18671
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New Failures, 3 Cancelled Jobs, 2 Unrelated FailuresAs of commit af3deb1 with merge base 38c5ca3 ( NEW FAILURES - The following jobs have failed:
CANCELLED JOBS - The following jobs were cancelled. Please retry:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
BROKEN TRUNK - The following job failed but was present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
|
@JorickvdHoeven has exported this pull request. If you are a Meta employee, you can view the originating Diff in D99177468. |
Summary: Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
5c6f96a to
cc1c74f
Compare
Summary: Pull Request resolved: #18671 Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
cc1c74f to
56ac3dc
Compare
Summary: Pull Request resolved: #18671 Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
56ac3dc to
1c90c12
Compare
Summary: Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
1c90c12 to
3d43271
Compare
Summary: Pull Request resolved: #18671 Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
3d43271 to
24d54e6
Compare
Summary: Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
24d54e6 to
2eef568
Compare
Summary: Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
2eef568 to
9214ce4
Compare
Summary: Pull Request resolved: #18671 Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
9214ce4 to
20c3688
Compare
Summary: Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
20c3688 to
ef487f5
Compare
Summary: Pull Request resolved: #18671 Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
ef487f5 to
5ffe283
Compare
Summary: Pull Request resolved: #18671 Adds quantized (INT) support for `torch.log10` in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in the `InsertTableOpsPass`. Changes: - Registered `log10.default` in the `unary_table_ops` dict in `insert_table_ops.py` for the LUT quantized path - Added `log10.default` to `_one_to_one` in `quantization_annotator.py` to enable quantization annotation - Added `log10.default` to `TOSA_PRO_INT_SupportList` in `tosa_profile_supported_op_lists.py` so the partitioner delegates it to the Arm backend - Wrote `test/ops/test_log10.py` with TOSA INT, U55 INT, and U85 INT test cases - Registered the test in `test/targets.bzl Reviewed By: 3l1 Differential Revision: D99177468
5ffe283 to
af3deb1
Compare
Summary:
Adds quantized (INT) support for
torch.log10in the ExecuTorch Arm backend using the lookup table (LUT) path. TOSA has no native LOG10 op, so for quantized inference the op is handled via a precomputed table in theInsertTableOpsPass.Changes:
log10.defaultin theunary_table_opsdict ininsert_table_ops.pyfor the LUT quantized pathlog10.defaultto_one_to_oneinquantization_annotator.pyto enable quantization annotationlog10.defaulttoTOSA_PRO_INT_SupportListintosa_profile_supported_op_lists.pyso the partitioner delegates it to the Arm backendtest/ops/test_log10.pywith TOSA INT, U55 INT, and U85 INT test casesReviewed By: 3l1
Differential Revision: D99177468