[IR] Remove mkdirs call in external tensor save #2033

justinchuby · 2025-01-23T04:47:51Z

The user should ensure the directory exists. This is consistent with the behavior of torch.export.save and other python methods.

Copilot

Copilot reviewed 1 out of 1 changed files in this pull request and generated no comments.

Comments suppressed due to low confidence (2)

onnxscript/ir/external_data.py:250

The condition 'if dir_path:' may not handle all edge cases correctly. Consider verifying if 'dir_path' can be None or other unexpected values. Ensure that this condition is appropriate and doesn't introduce any unintended behavior.

if dir_path:

onnxscript/ir/external_data.py:250

Ensure that there are tests covering the new behavior introduced by the condition 'if dir_path:'. This will help validate that the change works as intended and doesn't introduce any bugs.

if dir_path:

codecov · 2025-01-23T04:51:42Z

❌ 106 Tests Failed:

Tests completed	Failed	Passed	Skipped
11978	106	11872	2455

View the top 3 failed tests by shortest run time

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0243_test_clip_default_min

Stack Traces | 0.003s run time

onnxscript\backend\onnx_export_test.py:137: in extract_functions
    mod = importlib.import_module(import_name)
C:\hostedtoolcache\windows\Python\3.10.11\x64\lib\importlib\__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
E   ModuleNotFoundError: No module named 'tests.onnx_backend_test_code.test_clip_default_min'

The above exception was the direct cause of the following exception:
.nox\test\lib\site-packages\parameterized\parameterized.py:620: in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
onnxscript\backend\onnx_export_test.py:271: in test_export2python_produces_correct_onnx_script_model
    functions = extract_functions(backend_test.name, code, self.test_folder)
onnxscript\backend\onnx_export_test.py:139: in extract_functions
    raise AssertionError(
E   AssertionError: Unable to import 'tests.onnx_backend_test_code.test_clip_default_min' (e=No module named 'tests.onnx_backend_test_code.test_clip_default_min') (file: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_clip_default_min.py', absolute path: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_clip_default_min.py', current folder: D:\a\onnxscript\onnxscript
E   ---- CONTENT --
E   import numpy
E   from onnx import TensorProto
E   from onnx.helper import make_tensor
E   from onnxscript import script, external_tensor
E   from onnxscript.values import Opset
E   from onnxscript.onnx_types import FLOAT
E   from onnxscript.onnx_opset import opset13
E   
E   @script()
E   def bck_test_clip_default_min(x: FLOAT[3,4,5], min: FLOAT) -> (FLOAT[3,4,5]):
E       y = opset13.Clip(x, min)
E       return y

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0740_test_onehot_with_negative_axis

Stack Traces | 0.003s run time

onnxscript\backend\onnx_export_test.py:137: in extract_functions
    mod = importlib.import_module(import_name)
C:\hostedtoolcache\windows\Python\3.10.11\x64\lib\importlib\__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
E   ModuleNotFoundError: No module named 'tests.onnx_backend_test_code.test_onehot_with_negative_axis'

The above exception was the direct cause of the following exception:
.nox\test\lib\site-packages\parameterized\parameterized.py:620: in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
onnxscript\backend\onnx_export_test.py:271: in test_export2python_produces_correct_onnx_script_model
    functions = extract_functions(backend_test.name, code, self.test_folder)
onnxscript\backend\onnx_export_test.py:139: in extract_functions
    raise AssertionError(
E   AssertionError: Unable to import 'tests.onnx_backend_test_code.test_onehot_with_negative_axis' (e=No module named 'tests.onnx_backend_test_code.test_onehot_with_negative_axis') (file: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_onehot_with_negative_axis.py', absolute path: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_onehot_with_negative_axis.py', current folder: D:\a\onnxscript\onnxscript
E   ---- CONTENT --
E   import numpy
E   from onnx import TensorProto
E   from onnx.helper import make_tensor
E   from onnxscript import script, external_tensor
E   from onnxscript.values import Opset
E   from onnxscript.onnx_types import FLOAT
E   from onnxscript.onnx_opset import opset11
E   
E   @script()
E   def bck_test_onehot_with_negative_axis(indices: FLOAT[2,2], depth: FLOAT, values: FLOAT[2]) -> (FLOAT[2,10,2]):
E       y = opset11.OneHot(indices, depth, values, axis=-2)
E       return y

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0799_test_reciprocal

Stack Traces | 0.003s run time

onnxscript\backend\onnx_export_test.py:137: in extract_functions
    mod = importlib.import_module(import_name)
C:\hostedtoolcache\windows\Python\3.11.9\x64\Lib\importlib\__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
E   ModuleNotFoundError: No module named 'tests.onnx_backend_test_code.test_reciprocal'

The above exception was the direct cause of the following exception:
.nox\test_ort_nightly\Lib\site-packages\parameterized\parameterized.py:620: in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
onnxscript\backend\onnx_export_test.py:271: in test_export2python_produces_correct_onnx_script_model
    functions = extract_functions(backend_test.name, code, self.test_folder)
onnxscript\backend\onnx_export_test.py:139: in extract_functions
    raise AssertionError(
E   AssertionError: Unable to import 'tests.onnx_backend_test_code.test_reciprocal' (e=No module named 'tests.onnx_backend_test_code.test_reciprocal') (file: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_reciprocal.py', absolute path: 'D:\\a\\onnxscript\\onnxscript\\tests\\onnx_backend_test_code\\test_reciprocal.py', current folder: D:\a\onnxscript\onnxscript
E   ---- CONTENT --
E   import numpy
E   from onnx import TensorProto
E   from onnx.helper import make_tensor
E   from onnxscript import script, external_tensor
E   from onnxscript.values import Opset
E   from onnxscript.onnx_types import FLOAT
E   from onnxscript.onnx_opset import opset13
E   
E   @script()
E   def bck_test_reciprocal(x: FLOAT[3,4,5]) -> (FLOAT[3,4,5]):
E       y = opset13.Reciprocal(x)
E       return y

To view more test analytics, go to the Test Analytics Dashboard
📢 Thoughts on this report? Let us know!

Copilot

Copilot reviewed 1 out of 1 changed files in this pull request and generated no comments.

### Description This PR adds fusions for [Google's SigLIP model](https://huggingface.co/google/siglip-base-patch16-224/) and Microsoft's internal conformer-encoder model. Here is an example of how to run the ORT transformer optimizer for the SigLIP model. ``` $ git clone https://github.com/microsoft/onnxruntime $ cd onnxruntime/onnxruntime/python/tools/transformers $ python3 optimizer.py --input /path/to/model.onnx --output /path/to/model_opt.onnx --model_type clip --num_heads 16 --hidden_size 1152 --use_external_data_format --opt_level 0 --disable_shape_inference ``` Here is an example of how to run the ORT transformer optimizer for the conformer-encoder model. ``` $ git clone https://github.com/microsoft/onnxruntime $ cd onnxruntime/onnxruntime/python/tools/transformers $ python3 optimizer.py --input /path/to/model.onnx --output /path/to/model_opt.onnx --model_type conformer --num_heads 16 --hidden_size 1024 --use_external_data_format --opt_level 0 --disable_shape_inference --convert_attribute ``` ### Motivation and Context This PR helps optimize multi-modal models that use SigLIP for the vision encoder and conformer-encoder for the speech encoder. This PR uses changes from the following PRs: - pytorch/pytorch#144801 - microsoft/onnxscript#2018 - microsoft/onnxscript#2019 - microsoft/onnxscript#2020 - microsoft/onnxscript#2021 - microsoft/onnxscript#2022 - microsoft/onnxscript#2024 - microsoft/onnxscript#2025 - microsoft/onnxscript#2029 - microsoft/onnxscript#2033 ### Introduction of ONNX Script This PR introduces [ONNX Script](https://github.com/microsoft/onnxscript) into the ORT transformer optimizer as an optional step via the `fold_transpose_initializers()` method of the `DynamoOnnxHelper` class.

[IR] handle with external data path is empty or in the same dir as model

47bd05e

justinchuby requested a review from Copilot January 23, 2025 04:47

Copilot AI reviewed Jan 23, 2025

View reviewed changes

Remove call

14f2287

justinchuby changed the title ~~[IR] handle with external data path is empty or in the same dir as model~~ [IR] Remove mkdirs call in external tensor save Jan 23, 2025

justinchuby requested a review from Copilot January 23, 2025 16:08

justinchuby marked this pull request as ready for review January 23, 2025 16:08

Copilot AI reviewed Jan 23, 2025

View reviewed changes

justinchuby added merge at lgtm Reviewers can merge when they approve module: IR Intermediate representation labels Jan 23, 2025

justinchuby requested a review from titaiwangms January 23, 2025 16:09

justinchuby enabled auto-merge (squash) January 23, 2025 16:09

xadupre approved these changes Jan 23, 2025

View reviewed changes

Merge branch 'main' into justinchu/fix-external

fa8f9d1

justinchuby merged commit e673351 into main Jan 23, 2025
20 of 27 checks passed

justinchuby deleted the justinchu/fix-external branch January 23, 2025 18:41

kunal-vaishnavi mentioned this pull request Jan 29, 2025

Add fusions for SigLIP and Conformer-Encoder microsoft/onnxruntime#23528

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[IR] Remove mkdirs call in external tensor save #2033

[IR] Remove mkdirs call in external tensor save #2033

Uh oh!

justinchuby commented Jan 23, 2025 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

codecov bot commented Jan 23, 2025 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

[IR] Remove mkdirs call in external tensor save #2033

[IR] Remove mkdirs call in external tensor save #2033

Uh oh!

Conversation

justinchuby commented Jan 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

codecov bot commented Jan 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

❌ 106 Tests Failed:

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

justinchuby commented Jan 23, 2025 •

edited

Loading

codecov bot commented Jan 23, 2025 •

edited

Loading