[IR] Expose the Tape module #2127

justinchuby · 2025-03-24T20:01:28Z

Expose the Tape class under ir.tape for simplifying graph construction in the IR. This is a secondary API for convenience. I updated onnxscript/ir/passes/common/shape_inference_test.py to demonstrate usage.

I added an optional reference to the graph from Tape. When the graph is specified, the added nodes are appended to the graph. This provides users the ability to examine the graph as they build it up using Tape.

Copilot

Pull Request Overview

This PR exposes the Tape module as part of the public API for simplified graph construction in the IR and updates shape inference tests to demonstrate its usage.

Exposes and re-exports the Tape class under onnxscript/ir/tape.py and via onnxscript/ir/init.py.
Refactors the Tape class in onnxscript/ir/_tape.py to accept additional parameters for op methods.
Updates shape inference tests to use the new Tape API instead of directly instantiating Node objects.

Reviewed Changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
onnxscript/ir/tape.py	Introduces the public API for the Tape class and sets its module.
onnxscript/ir/init.py	Exports the tape module to be part of the IR package’s public API.
onnxscript/ir/_tape.py	Updates the Tape class and op methods with additional parameters.
onnxscript/ir/passes/common/shape_inference_test.py	Revises tests to utilize the Tape API for constructing graphs.

Comments suppressed due to low confidence (2)

onnxscript/ir/_tape.py:99

In op_multi_output, the 'num_outputs' parameter is now optional and may be None. Ensure that the Node constructor and downstream logic can handle a None value or consider providing a default integer value.

num_outputs: int | None = None,

onnxscript/ir/passes/common/shape_inference_test.py:73

[nitpick] Consider using the 'inputs' list directly instead of '[*inputs]' for consistency with other tests and to reduce unnecessary overhead.

val_add = tape.op("Add", inputs=[*inputs])

Copilot

Pull Request Overview

This PR exposes the Tape module to simplify IR graph construction by providing a convenient, secondary API. Key changes include:

Introducing onnxscript/ir/tape.py to export the Tape class.
Updating onnxscript/ir/_tape.py with extended op and op_multi_output signatures.
Modifying tests in onnxscript/ir/passes/common/shape_inference_test.py to demonstrate Tape usage.

Reviewed Changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
onnxscript/ir/tape.py	Added public API for the Tape module and set the module name.
onnxscript/ir/_tape.py	Extended the Tape class with additional parameters in methods.
onnxscript/ir/init.py	Updated exports to include the Tape module.
onnxscript/ir/passes/common/shape_inference_test.py	Updated tests to use the new Tape API for node operations.

codecov · 2025-03-24T20:06:46Z

❌ 5 Tests Failed:

Tests completed	Failed	Passed	Skipped
15603	5	15598	2741

View the top 3 failed test(s) by shortest run time

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0125_test_ai_onnx_ml_tree_ensemble_set_membership

Stack Traces | 0.007s run time

onnxscript/converter.py:467: in _eval_constant_expr
    return eval(cpl, self.globals, locals)  # pylint: disable=eval-used
E   NameError: name 'nan' is not defined

The above exception was the direct cause of the following exception:
..../test_ort_nightly/lib/python3.11.../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:137: in extract_functions
    mod = importlib.import_module(import_name)
.../Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/importlib/__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
<frozen importlib._bootstrap>:1204: in _gcd_import
    ???
<frozen importlib._bootstrap>:1176: in _find_and_load
    ???
<frozen importlib._bootstrap>:1147: in _find_and_load_unlocked
    ???
<frozen importlib._bootstrap>:690: in _load_unlocked
    ???
..../test_ort_nightly/lib/python3.11.../_pytest/assertion/rewrite.py:185: in exec_module
    exec(co, module.__dict__)
tests/onnx_backend_test_code/test_ai_onnx_ml_tree_ensemble_set_membership.py:9: in <module>
    @script()
onnxscript/main.py:91: in transform
    result = script_check(f_ast, opset, env, src, default_opset=default_opset)
onnxscript/main.py:35: in script_check
    return convert.translate_function_def(f)
onnxscript/converter.py:1459: in translate_function_def
    fn_ir = self._translate_function_def_common(stmt)
onnxscript/converter.py:1446: in _translate_function_def_common
    self._translate_stmt(s, index_of_stmt=i)
onnxscript/converter.py:968: in _translate_stmt
    return self._translate_assign_stmt(node)
onnxscript/converter.py:1055: in _translate_assign_stmt
    assign(lhs, rhs)
onnxscript/converter.py:999: in assign
    t = self._translate_expr(rhs, lhs).name
onnxscript/converter.py:553: in _translate_expr
    r = self._translate_call_expr(node)
onnxscript/converter.py:832: in _translate_call_expr
    attrs = [
onnxscript/converter.py:833: in <listcomp>
    self._translate_attr(x, y, callee.op_schema.attributes[x])
onnxscript/converter.py:517: in _translate_attr
    val = self._eval_constant_expr(expr)
onnxscript/converter.py:469: in _eval_constant_expr
    raise NameError(
E   NameError: ERROR: Missing names, globals contains ['__name__', '__doc__', '__package__', '__loader__', '__spec__', '__file__', '__cached__', '__builtins__', '@py_builtins', '@pytest_ar', 'numpy', 'TensorProto', 'make_tensor', 'script', 'external_tensor', 'Opset', 'FLOAT', 'ai_onnx_ml5'], locals [].
E   at: Function 'bck_test_ai_onnx_ml_tree_ensemble_set_membership', line 3
E       Y = ai_onnx_ml5.TreeEnsemble(X, aggregate_function=1, leaf_targetids=[0, 1, 2, 3], leaf_weights=make_tensor("value", 1, dims=[4], vals=[1.0, 10.0, 1000.0, 100.0]), membership_values=make_tensor("value", 1, dims=[8], vals=[1.2000000476837158, 3.700000047683716, 8.0, 9.0, nan, 12.0, 7.0, nan]), n_targets=4, nodes_falseleafs=[1, 0, 1], nodes_falsenodeids=[2, 2, 3], nodes_featureids=[0, 0, 0], nodes_modes=make_tensor("value", 2, dims=[3], vals=[0, 6, 6]), nodes_splits=make_tensor("value", 1, dims=[3], vals=[11.0, 232344.0, nan]), nodes_trueleafs=[0, 1, 1], nodes_truenodeids=[1, 0, 1], post_transform=0, tree_roots=[0])
E                                                                                                                                                                                             ^

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0397_test_ai_onnx_ml_tree_ensemble_set_membership

Stack Traces | 0.01s run time

onnxscript/converter.py:467: in _eval_constant_expr
    return eval(cpl, self.globals, locals)  # pylint: disable=eval-used
E   NameError: name 'nan' is not defined

The above exception was the direct cause of the following exception:
..../test_ort_nightly/lib/python3.11.../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:137: in extract_functions
    mod = importlib.import_module(import_name)
.../hostedtoolcache/Python/3.11.11.../x64/lib/python3.11/importlib/__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
<frozen importlib._bootstrap>:1204: in _gcd_import
    ???
<frozen importlib._bootstrap>:1176: in _find_and_load
    ???
<frozen importlib._bootstrap>:1147: in _find_and_load_unlocked
    ???
<frozen importlib._bootstrap>:690: in _load_unlocked
    ???
..../test_ort_nightly/lib/python3.11.../_pytest/assertion/rewrite.py:185: in exec_module
    exec(co, module.__dict__)
tests/onnx_backend_test_code/test_ai_onnx_ml_tree_ensemble_set_membership.py:9: in <module>
    @script()
onnxscript/main.py:91: in transform
    result = script_check(f_ast, opset, env, src, default_opset=default_opset)
onnxscript/main.py:35: in script_check
    return convert.translate_function_def(f)
onnxscript/converter.py:1459: in translate_function_def
    fn_ir = self._translate_function_def_common(stmt)
onnxscript/converter.py:1446: in _translate_function_def_common
    self._translate_stmt(s, index_of_stmt=i)
onnxscript/converter.py:968: in _translate_stmt
    return self._translate_assign_stmt(node)
onnxscript/converter.py:1055: in _translate_assign_stmt
    assign(lhs, rhs)
onnxscript/converter.py:999: in assign
    t = self._translate_expr(rhs, lhs).name
onnxscript/converter.py:553: in _translate_expr
    r = self._translate_call_expr(node)
onnxscript/converter.py:832: in _translate_call_expr
    attrs = [
onnxscript/converter.py:833: in <listcomp>
    self._translate_attr(x, y, callee.op_schema.attributes[x])
onnxscript/converter.py:517: in _translate_attr
    val = self._eval_constant_expr(expr)
onnxscript/converter.py:469: in _eval_constant_expr
    raise NameError(
E   NameError: ERROR: Missing names, globals contains ['__name__', '__doc__', '__package__', '__loader__', '__spec__', '__file__', '__cached__', '__builtins__', '@py_builtins', '@pytest_ar', 'numpy', 'TensorProto', 'make_tensor', 'script', 'external_tensor', 'Opset', 'FLOAT', 'ai_onnx_ml5'], locals [].
E   at: Function 'bck_test_ai_onnx_ml_tree_ensemble_set_membership', line 3
E       Y = ai_onnx_ml5.TreeEnsemble(X, aggregate_function=1, leaf_targetids=[0, 1, 2, 3], leaf_weights=make_tensor("value", 1, dims=[4], vals=[1.0, 10.0, 1000.0, 100.0]), membership_values=make_tensor("value", 1, dims=[8], vals=[1.2000000476837158, 3.700000047683716, 8.0, 9.0, nan, 12.0, 7.0, nan]), n_targets=4, nodes_falseleafs=[1, 0, 1], nodes_falsenodeids=[2, 2, 3], nodes_featureids=[0, 0, 0], nodes_modes=make_tensor("value", 2, dims=[3], vals=[0, 6, 6]), nodes_splits=make_tensor("value", 1, dims=[3], vals=[11.0, 232344.0, nan]), nodes_trueleafs=[0, 1, 1], nodes_truenodeids=[1, 0, 1], post_transform=0, tree_roots=[0])
E                                                                                                                                                                                             ^

onnxscript.backend.onnx_export_test.TestOnnxBackEnd::test_export2python_produces_correct_onnx_script_model_0026_test_ai_onnx_ml_tree_ensemble_set_membership

Stack Traces | 0.048s run time

onnxscript\converter.py:467: in _eval_constant_expr
    return eval(cpl, self.globals, locals)  # pylint: disable=eval-used
E   NameError: name 'nan' is not defined

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: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)
<frozen importlib._bootstrap>:1204: in _gcd_import
    ???
<frozen importlib._bootstrap>:1176: in _find_and_load
    ???
<frozen importlib._bootstrap>:1147: in _find_and_load_unlocked
    ???
<frozen importlib._bootstrap>:690: in _load_unlocked
    ???
.nox\test_ort_nightly\Lib\site-packages\_pytest\assertion\rewrite.py:185: in exec_module
    exec(co, module.__dict__)
tests\onnx_backend_test_code\test_ai_onnx_ml_tree_ensemble_set_membership.py:9: in <module>
    @script()
onnxscript\main.py:91: in transform
    result = script_check(f_ast, opset, env, src, default_opset=default_opset)
onnxscript\main.py:35: in script_check
    return convert.translate_function_def(f)
onnxscript\converter.py:1459: in translate_function_def
    fn_ir = self._translate_function_def_common(stmt)
onnxscript\converter.py:1446: in _translate_function_def_common
    self._translate_stmt(s, index_of_stmt=i)
onnxscript\converter.py:968: in _translate_stmt
    return self._translate_assign_stmt(node)
onnxscript\converter.py:1055: in _translate_assign_stmt
    assign(lhs, rhs)
onnxscript\converter.py:999: in assign
    t = self._translate_expr(rhs, lhs).name
onnxscript\converter.py:553: in _translate_expr
    r = self._translate_call_expr(node)
onnxscript\converter.py:832: in _translate_call_expr
    attrs = [
onnxscript\converter.py:833: in <listcomp>
    self._translate_attr(x, y, callee.op_schema.attributes[x])
onnxscript\converter.py:517: in _translate_attr
    val = self._eval_constant_expr(expr)
onnxscript\converter.py:469: in _eval_constant_expr
    raise NameError(
E   NameError: ERROR: Missing names, globals contains ['__name__', '__doc__', '__package__', '__loader__', '__spec__', '__file__', '__cached__', '__builtins__', '@py_builtins', '@pytest_ar', 'numpy', 'TensorProto', 'make_tensor', 'script', 'external_tensor', 'Opset', 'FLOAT', 'ai_onnx_ml5'], locals [].
E   at: Function 'bck_test_ai_onnx_ml_tree_ensemble_set_membership', line 3
E       Y = ai_onnx_ml5.TreeEnsemble(X, aggregate_function=1, leaf_targetids=[0, 1, 2, 3], leaf_weights=make_tensor("value", 1, dims=[4], vals=[1.0, 10.0, 1000.0, 100.0]), membership_values=make_tensor("value", 1, dims=[8], vals=[1.2000000476837158, 3.700000047683716, 8.0, 9.0, nan, 12.0, 7.0, nan]), n_targets=4, nodes_falseleafs=[1, 0, 1], nodes_falsenodeids=[2, 2, 3], nodes_featureids=[0, 0, 0], nodes_modes=make_tensor("value", 2, dims=[3], vals=[0, 6, 6]), nodes_splits=make_tensor("value", 1, dims=[3], vals=[11.0, 232344.0, nan]), nodes_trueleafs=[0, 1, 1], nodes_truenodeids=[1, 0, 1], post_transform=0, tree_roots=[0])
E                                                                                                                                                                                             ^

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}

onnxscript/ir/_tape.py

onnxscript/ir/passes/common/shape_inference_test.py

onnxscript/ir/_tape.py

Copilot

Pull Request Overview

This PR exposes the Tape module to simplify graph construction in the IR while updating tests to demonstrate and support the new API.

Introduces onnxscript/ir/tape.py exposing the Tape class.
Updates shape inference and initializer tests to use Tape operations.
Extends the Tape API (and Builder subclass) and updates core type hints to support both Graph and Function types.

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
onnxscript/ir/tape.py	Exposes the Tape class with a public API and sets its module.
onnxscript/ir/passes/common/shape_inference_test.py	Updates tests to use Tape for creating operations and nodes.
onnxscript/ir/_tape_test.py	Adds tests covering tape operations, initializers, and multi-output.
onnxscript/ir/_tape.py	Extends the Tape API to include additional parameters and tracks used opsets.
onnxscript/ir/_core.py	Adjusts type hints for the graph parameter to accept Function objects.
onnxscript/ir/init.py	Includes the tape module in the public IR API exports.

onnxscript/ir/_tape_test.py

Copilot

Pull Request Overview

This PR exposes the Tape module as part of the public IR API for simplified graph construction and improved testing.

Introduces onnxscript/ir/tape.py to expose the Tape class.
Updates tests in onnxscript/ir/passes/common/shape_inference_test.py and onnxscript/ir/_tape_test.py to demonstrate Tape usage.
Enhances the internal Tape implementation and type hints in onnxscript/ir/_tape.py and onnxscript/ir/_core.py to support both Graph and Function types.

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
onnxscript/ir/tape.py	Introduces public Tape API by importing from _tape and exposing it under the ir.tape namespace.
onnxscript/ir/passes/common/shape_inference_test.py	Updates tests to utilize Tape for node creation and graph construction.
onnxscript/ir/_tape_test.py	Adds new tests for Tape functionality including operations and multi-output handling.
onnxscript/ir/_tape.py	Extends Tape implementation with additional parameters, type hints, and used_opsets management.
onnxscript/ir/_core.py	Updates type hints for the graph parameter to include support for Function objects.
onnxscript/ir/init.py	Exports the new tape module as part of the public IR API.

shubhambhokare1

Thanks, nit: lint and rebase

justinchuby added 3 commits March 24, 2025 11:57

tape

573ada3

Create a public tape

903e48e

tape

627d30f

github-project-automation bot added this to ONNX Script Review Board Mar 24, 2025

github-project-automation bot moved this to Todo in ONNX Script Review Board Mar 24, 2025

justinchuby requested a review from Copilot March 24, 2025 20:01

Copilot AI reviewed Mar 24, 2025

View reviewed changes

fix tests

1a43ace

justinchuby requested a review from Copilot March 24, 2025 20:06

Copilot AI reviewed Mar 24, 2025

View reviewed changes

inputs

1031e42

num_outputs

fad3122

justinchuby added the module: IR label Mar 24, 2025

shubhambhokare1 reviewed Mar 24, 2025

View reviewed changes

onnxscript/ir/_tape.py Outdated Show resolved Hide resolved

onnxscript/ir/passes/common/shape_inference_test.py Show resolved Hide resolved

justinchuby added 2 commits March 24, 2025 17:11

Merge branch 'main' into justinchu/tape

0162f36

example

4e6284c

xadupre reviewed Mar 25, 2025

View reviewed changes

onnxscript/ir/_tape.py Show resolved Hide resolved

xadupre reviewed Mar 25, 2025

View reviewed changes

onnxscript/ir/_tape.py Show resolved Hide resolved

justinchuby added 4 commits March 28, 2025 09:56

Merge branch 'main' into justinchu/tape

4292da7

used_opsets

224c86f

Take a graph

86266bd

test

1741b71

justinchuby requested review from gramalingam, shubhambhokare1, xadupre and Copilot March 28, 2025 17:20

Copilot AI reviewed Mar 28, 2025

View reviewed changes

Merge branch 'main' into justinchu/tape

cd99b2a

github-advanced-security bot found potential problems Mar 28, 2025

View reviewed changes

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

github-advanced-security bot found potential problems Mar 28, 2025

View reviewed changes

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

onnxscript/ir/_tape_test.py Fixed Show fixed Hide fixed

lint

f6120f0

justinchuby requested a review from Copilot March 31, 2025 18:42

Copilot AI reviewed Mar 31, 2025

View reviewed changes

shubhambhokare1 previously approved these changes Apr 7, 2025

View reviewed changes

justinchuby added 3 commits April 8, 2025 09:30

Merge branch 'main' into justinchu/tape

acfa81c

typing

081ae34

type

1745761

justinchuby dismissed shubhambhokare1’s stale review via 1745761 April 8, 2025 16:39

format

f2da634

justinchuby enabled auto-merge (squash) April 8, 2025 16:43

shubhambhokare1 approved these changes Apr 8, 2025

View reviewed changes

justinchuby merged commit ad64b58 into main Apr 8, 2025
21 of 27 checks passed

justinchuby deleted the justinchu/tape branch April 8, 2025 17:00

github-project-automation bot moved this from Todo to Done in ONNX Script Review Board Apr 8, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[IR] Expose the Tape module #2127

[IR] Expose the Tape module #2127

justinchuby commented Mar 24, 2025 •

edited

Loading

Copilot AI left a comment

Copilot AI left a comment

codecov bot commented Mar 24, 2025 •

edited

Loading

Copilot AI left a comment

Copilot AI left a comment

shubhambhokare1 left a comment •

edited

Loading

[IR] Expose the Tape module #2127

[IR] Expose the Tape module #2127

Conversation

justinchuby commented Mar 24, 2025 • edited Loading

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

codecov bot commented Mar 24, 2025 • edited Loading

❌ 5 Tests Failed:

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

shubhambhokare1 left a comment • edited Loading

Choose a reason for hiding this comment

justinchuby commented Mar 24, 2025 •

edited

Loading

codecov bot commented Mar 24, 2025 •

edited

Loading

shubhambhokare1 left a comment •

edited

Loading