mirror of
https://github.com/open-telemetry/opentelemetry-python-contrib.git
synced 2025-08-01 17:34:38 +08:00
Merge pull request #151 from crflynn/opentelemetry-instrumentation-sklearn
This commit is contained in:
Leighton Chen
@ -0,0 +1,5 @@
|
|||||||
|
# Changelog
|
||||||
|
|
||||||
|
## Unreleased
|
||||||
|
|
||||||
|
- Initial release ([#151](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/151))
|
||||||
201
instrumentation/opentelemetry-instrumentation-sklearn/LICENSE
Normal file
201
instrumentation/opentelemetry-instrumentation-sklearn/LICENSE
Normal file
@ -0,0 +1,201 @@
|
|||||||
|
Apache License
|
||||||
|
Version 2.0, January 2004
|
||||||
|
http://www.apache.org/licenses/
|
||||||
|
|
||||||
|
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
||||||
|
|
||||||
|
1. Definitions.
|
||||||
|
|
||||||
|
"License" shall mean the terms and conditions for use, reproduction,
|
||||||
|
and distribution as defined by Sections 1 through 9 of this document.
|
||||||
|
|
||||||
|
"Licensor" shall mean the copyright owner or entity authorized by
|
||||||
|
the copyright owner that is granting the License.
|
||||||
|
|
||||||
|
"Legal Entity" shall mean the union of the acting entity and all
|
||||||
|
other entities that control, are controlled by, or are under common
|
||||||
|
control with that entity. For the purposes of this definition,
|
||||||
|
"control" means (i) the power, direct or indirect, to cause the
|
||||||
|
direction or management of such entity, whether by contract or
|
||||||
|
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
||||||
|
outstanding shares, or (iii) beneficial ownership of such entity.
|
||||||
|
|
||||||
|
"You" (or "Your") shall mean an individual or Legal Entity
|
||||||
|
exercising permissions granted by this License.
|
||||||
|
|
||||||
|
"Source" form shall mean the preferred form for making modifications,
|
||||||
|
including but not limited to software source code, documentation
|
||||||
|
source, and configuration files.
|
||||||
|
|
||||||
|
"Object" form shall mean any form resulting from mechanical
|
||||||
|
transformation or translation of a Source form, including but
|
||||||
|
not limited to compiled object code, generated documentation,
|
||||||
|
and conversions to other media types.
|
||||||
|
|
||||||
|
"Work" shall mean the work of authorship, whether in Source or
|
||||||
|
Object form, made available under the License, as indicated by a
|
||||||
|
copyright notice that is included in or attached to the work
|
||||||
|
(an example is provided in the Appendix below).
|
||||||
|
|
||||||
|
"Derivative Works" shall mean any work, whether in Source or Object
|
||||||
|
form, that is based on (or derived from) the Work and for which the
|
||||||
|
editorial revisions, annotations, elaborations, or other modifications
|
||||||
|
represent, as a whole, an original work of authorship. For the purposes
|
||||||
|
of this License, Derivative Works shall not include works that remain
|
||||||
|
separable from, or merely link (or bind by name) to the interfaces of,
|
||||||
|
the Work and Derivative Works thereof.
|
||||||
|
|
||||||
|
"Contribution" shall mean any work of authorship, including
|
||||||
|
the original version of the Work and any modifications or additions
|
||||||
|
to that Work or Derivative Works thereof, that is intentionally
|
||||||
|
submitted to Licensor for inclusion in the Work by the copyright owner
|
||||||
|
or by an individual or Legal Entity authorized to submit on behalf of
|
||||||
|
the copyright owner. For the purposes of this definition, "submitted"
|
||||||
|
means any form of electronic, verbal, or written communication sent
|
||||||
|
to the Licensor or its representatives, including but not limited to
|
||||||
|
communication on electronic mailing lists, source code control systems,
|
||||||
|
and issue tracking systems that are managed by, or on behalf of, the
|
||||||
|
Licensor for the purpose of discussing and improving the Work, but
|
||||||
|
excluding communication that is conspicuously marked or otherwise
|
||||||
|
designated in writing by the copyright owner as "Not a Contribution."
|
||||||
|
|
||||||
|
"Contributor" shall mean Licensor and any individual or Legal Entity
|
||||||
|
on behalf of whom a Contribution has been received by Licensor and
|
||||||
|
subsequently incorporated within the Work.
|
||||||
|
|
||||||
|
2. Grant of Copyright License. Subject to the terms and conditions of
|
||||||
|
this License, each Contributor hereby grants to You a perpetual,
|
||||||
|
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
||||||
|
copyright license to reproduce, prepare Derivative Works of,
|
||||||
|
publicly display, publicly perform, sublicense, and distribute the
|
||||||
|
Work and such Derivative Works in Source or Object form.
|
||||||
|
|
||||||
|
3. Grant of Patent License. Subject to the terms and conditions of
|
||||||
|
this License, each Contributor hereby grants to You a perpetual,
|
||||||
|
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
||||||
|
(except as stated in this section) patent license to make, have made,
|
||||||
|
use, offer to sell, sell, import, and otherwise transfer the Work,
|
||||||
|
where such license applies only to those patent claims licensable
|
||||||
|
by such Contributor that are necessarily infringed by their
|
||||||
|
Contribution(s) alone or by combination of their Contribution(s)
|
||||||
|
with the Work to which such Contribution(s) was submitted. If You
|
||||||
|
institute patent litigation against any entity (including a
|
||||||
|
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
||||||
|
or a Contribution incorporated within the Work constitutes direct
|
||||||
|
or contributory patent infringement, then any patent licenses
|
||||||
|
granted to You under this License for that Work shall terminate
|
||||||
|
as of the date such litigation is filed.
|
||||||
|
|
||||||
|
4. Redistribution. You may reproduce and distribute copies of the
|
||||||
|
Work or Derivative Works thereof in any medium, with or without
|
||||||
|
modifications, and in Source or Object form, provided that You
|
||||||
|
meet the following conditions:
|
||||||
|
|
||||||
|
(a) You must give any other recipients of the Work or
|
||||||
|
Derivative Works a copy of this License; and
|
||||||
|
|
||||||
|
(b) You must cause any modified files to carry prominent notices
|
||||||
|
stating that You changed the files; and
|
||||||
|
|
||||||
|
(c) You must retain, in the Source form of any Derivative Works
|
||||||
|
that You distribute, all copyright, patent, trademark, and
|
||||||
|
attribution notices from the Source form of the Work,
|
||||||
|
excluding those notices that do not pertain to any part of
|
||||||
|
the Derivative Works; and
|
||||||
|
|
||||||
|
(d) If the Work includes a "NOTICE" text file as part of its
|
||||||
|
distribution, then any Derivative Works that You distribute must
|
||||||
|
include a readable copy of the attribution notices contained
|
||||||
|
within such NOTICE file, excluding those notices that do not
|
||||||
|
pertain to any part of the Derivative Works, in at least one
|
||||||
|
of the following places: within a NOTICE text file distributed
|
||||||
|
as part of the Derivative Works; within the Source form or
|
||||||
|
documentation, if provided along with the Derivative Works; or,
|
||||||
|
within a display generated by the Derivative Works, if and
|
||||||
|
wherever such third-party notices normally appear. The contents
|
||||||
|
of the NOTICE file are for informational purposes only and
|
||||||
|
do not modify the License. You may add Your own attribution
|
||||||
|
notices within Derivative Works that You distribute, alongside
|
||||||
|
or as an addendum to the NOTICE text from the Work, provided
|
||||||
|
that such additional attribution notices cannot be construed
|
||||||
|
as modifying the License.
|
||||||
|
|
||||||
|
You may add Your own copyright statement to Your modifications and
|
||||||
|
may provide additional or different license terms and conditions
|
||||||
|
for use, reproduction, or distribution of Your modifications, or
|
||||||
|
for any such Derivative Works as a whole, provided Your use,
|
||||||
|
reproduction, and distribution of the Work otherwise complies with
|
||||||
|
the conditions stated in this License.
|
||||||
|
|
||||||
|
5. Submission of Contributions. Unless You explicitly state otherwise,
|
||||||
|
any Contribution intentionally submitted for inclusion in the Work
|
||||||
|
by You to the Licensor shall be under the terms and conditions of
|
||||||
|
this License, without any additional terms or conditions.
|
||||||
|
Notwithstanding the above, nothing herein shall supersede or modify
|
||||||
|
the terms of any separate license agreement you may have executed
|
||||||
|
with Licensor regarding such Contributions.
|
||||||
|
|
||||||
|
6. Trademarks. This License does not grant permission to use the trade
|
||||||
|
names, trademarks, service marks, or product names of the Licensor,
|
||||||
|
except as required for reasonable and customary use in describing the
|
||||||
|
origin of the Work and reproducing the content of the NOTICE file.
|
||||||
|
|
||||||
|
7. Disclaimer of Warranty. Unless required by applicable law or
|
||||||
|
agreed to in writing, Licensor provides the Work (and each
|
||||||
|
Contributor provides its Contributions) on an "AS IS" BASIS,
|
||||||
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
||||||
|
implied, including, without limitation, any warranties or conditions
|
||||||
|
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
||||||
|
PARTICULAR PURPOSE. You are solely responsible for determining the
|
||||||
|
appropriateness of using or redistributing the Work and assume any
|
||||||
|
risks associated with Your exercise of permissions under this License.
|
||||||
|
|
||||||
|
8. Limitation of Liability. In no event and under no legal theory,
|
||||||
|
whether in tort (including negligence), contract, or otherwise,
|
||||||
|
unless required by applicable law (such as deliberate and grossly
|
||||||
|
negligent acts) or agreed to in writing, shall any Contributor be
|
||||||
|
liable to You for damages, including any direct, indirect, special,
|
||||||
|
incidental, or consequential damages of any character arising as a
|
||||||
|
result of this License or out of the use or inability to use the
|
||||||
|
Work (including but not limited to damages for loss of goodwill,
|
||||||
|
work stoppage, computer failure or malfunction, or any and all
|
||||||
|
other commercial damages or losses), even if such Contributor
|
||||||
|
has been advised of the possibility of such damages.
|
||||||
|
|
||||||
|
9. Accepting Warranty or Additional Liability. While redistributing
|
||||||
|
the Work or Derivative Works thereof, You may choose to offer,
|
||||||
|
and charge a fee for, acceptance of support, warranty, indemnity,
|
||||||
|
or other liability obligations and/or rights consistent with this
|
||||||
|
License. However, in accepting such obligations, You may act only
|
||||||
|
on Your own behalf and on Your sole responsibility, not on behalf
|
||||||
|
of any other Contributor, and only if You agree to indemnify,
|
||||||
|
defend, and hold each Contributor harmless for any liability
|
||||||
|
incurred by, or claims asserted against, such Contributor by reason
|
||||||
|
of your accepting any such warranty or additional liability.
|
||||||
|
|
||||||
|
END OF TERMS AND CONDITIONS
|
||||||
|
|
||||||
|
APPENDIX: How to apply the Apache License to your work.
|
||||||
|
|
||||||
|
To apply the Apache License to your work, attach the following
|
||||||
|
boilerplate notice, with the fields enclosed by brackets "[]"
|
||||||
|
replaced with your own identifying information. (Don't include
|
||||||
|
the brackets!) The text should be enclosed in the appropriate
|
||||||
|
comment syntax for the file format. We also recommend that a
|
||||||
|
file or class name and description of purpose be included on the
|
||||||
|
same "printed page" as the copyright notice for easier
|
||||||
|
identification within third-party archives.
|
||||||
|
|
||||||
|
Copyright [yyyy] [name of copyright owner]
|
||||||
|
|
||||||
|
Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
you may not use this file except in compliance with the License.
|
||||||
|
You may obtain a copy of the License at
|
||||||
|
|
||||||
|
http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
|
||||||
|
Unless required by applicable law or agreed to in writing, software
|
||||||
|
distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
See the License for the specific language governing permissions and
|
||||||
|
limitations under the License.
|
||||||
@ -0,0 +1,9 @@
|
|||||||
|
graft src
|
||||||
|
graft tests
|
||||||
|
global-exclude *.pyc
|
||||||
|
global-exclude *.pyo
|
||||||
|
global-exclude __pycache__/*
|
||||||
|
include CHANGELOG.md
|
||||||
|
include MANIFEST.in
|
||||||
|
include README.rst
|
||||||
|
include LICENSE
|
||||||
@ -0,0 +1,23 @@
|
|||||||
|
OpenTelemetry Scikit-Learn Instrumentation
|
||||||
|
==========================================
|
||||||
|
|
||||||
|
|pypi|
|
||||||
|
|
||||||
|
.. |pypi| image:: https://badge.fury.io/py/opentelemetry-instrumentation-sklearn.svg
|
||||||
|
:target: https://pypi.org/project/opentelemetry-instrumentation-sklearn/
|
||||||
|
|
||||||
|
This library allows tracing HTTP requests made by the
|
||||||
|
`scikit-learn <https://scikit-learn.org/stable/>`_ library.
|
||||||
|
|
||||||
|
Installation
|
||||||
|
------------
|
||||||
|
|
||||||
|
::
|
||||||
|
|
||||||
|
pip install opentelemetry-instrumentation-sklearn
|
||||||
|
|
||||||
|
References
|
||||||
|
----------
|
||||||
|
|
||||||
|
* `OpenTelemetry sklearn Instrumentation <https://opentelemetry-python.readthedocs.io/en/latest/instrumentation/sklearn/sklearn.html>`_
|
||||||
|
* `OpenTelemetry Project <https://opentelemetry.io/>`_
|
||||||
@ -0,0 +1,55 @@
|
|||||||
|
# Copyright The OpenTelemetry Authors
|
||||||
|
#
|
||||||
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
# you may not use this file except in compliance with the License.
|
||||||
|
# You may obtain a copy of the License at
|
||||||
|
#
|
||||||
|
# http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
#
|
||||||
|
# Unless required by applicable law or agreed to in writing, software
|
||||||
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
# See the License for the specific language governing permissions and
|
||||||
|
# limitations under the License.
|
||||||
|
#
|
||||||
|
[metadata]
|
||||||
|
name = opentelemetry-instrumentation-sklearn
|
||||||
|
description = OpenTelemetry sklearn instrumentation
|
||||||
|
long_description = file: README.rst
|
||||||
|
long_description_content_type = text/x-rst
|
||||||
|
author = OpenTelemetry Authors
|
||||||
|
author_email = cncf-opentelemetry-contributors@lists.cncf.io
|
||||||
|
url = https://github.com/open-telemetry/opentelemetry-python-contrib/tree/master/instrumentation/opentelemetry-instrumentation-sklearn
|
||||||
|
platforms = any
|
||||||
|
license = Apache-2.0
|
||||||
|
classifiers =
|
||||||
|
Development Status :: 4 - Beta
|
||||||
|
Intended Audience :: Developers
|
||||||
|
License :: OSI Approved :: Apache Software License
|
||||||
|
Programming Language :: Python
|
||||||
|
Programming Language :: Python :: 3
|
||||||
|
Programming Language :: Python :: 3.5
|
||||||
|
Programming Language :: Python :: 3.6
|
||||||
|
Programming Language :: Python :: 3.7
|
||||||
|
Programming Language :: Python :: 3.8
|
||||||
|
|
||||||
|
[options]
|
||||||
|
python_requires = >=3.5
|
||||||
|
package_dir=
|
||||||
|
=src
|
||||||
|
packages=find_namespace:
|
||||||
|
install_requires =
|
||||||
|
opentelemetry-api == 0.16.dev0
|
||||||
|
opentelemetry-instrumentation == 0.16.dev0
|
||||||
|
scikit-learn ~= 0.22.0
|
||||||
|
|
||||||
|
[options.extras_require]
|
||||||
|
test =
|
||||||
|
opentelemetry-test == 0.16.dev0
|
||||||
|
|
||||||
|
[options.packages.find]
|
||||||
|
where = src
|
||||||
|
|
||||||
|
[options.entry_points]
|
||||||
|
opentelemetry_instrumentor =
|
||||||
|
sklearn = opentelemetry.instrumentation.sklearn:SklearnInstrumentor
|
||||||
@ -0,0 +1,31 @@
|
|||||||
|
# Copyright The OpenTelemetry Authors
|
||||||
|
#
|
||||||
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
# you may not use this file except in compliance with the License.
|
||||||
|
# You may obtain a copy of the License at
|
||||||
|
#
|
||||||
|
# http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
#
|
||||||
|
# Unless required by applicable law or agreed to in writing, software
|
||||||
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
# See the License for the specific language governing permissions and
|
||||||
|
# limitations under the License.
|
||||||
|
import os
|
||||||
|
|
||||||
|
import setuptools
|
||||||
|
|
||||||
|
BASE_DIR = os.path.dirname(__file__)
|
||||||
|
VERSION_FILENAME = os.path.join(
|
||||||
|
BASE_DIR,
|
||||||
|
"src",
|
||||||
|
"opentelemetry",
|
||||||
|
"instrumentation",
|
||||||
|
"sklearn",
|
||||||
|
"version.py",
|
||||||
|
)
|
||||||
|
PACKAGE_INFO = {}
|
||||||
|
with open(VERSION_FILENAME) as f:
|
||||||
|
exec(f.read(), PACKAGE_INFO)
|
||||||
|
|
||||||
|
setuptools.setup(version=PACKAGE_INFO["__version__"])
|
||||||
@ -0,0 +1,759 @@
|
|||||||
|
# Copyright 2020, OpenTelemetry Authors
|
||||||
|
#
|
||||||
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
# you may not use this file except in compliance with the License.
|
||||||
|
# You may obtain a copy of the License at
|
||||||
|
#
|
||||||
|
# http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
#
|
||||||
|
# Unless required by applicable law or agreed to in writing, software
|
||||||
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
# See the License for the specific language governing permissions and
|
||||||
|
# limitations under the License.
|
||||||
|
"""
|
||||||
|
The integration with sklearn supports the scikit-learn compatible libraries,
|
||||||
|
it can be enabled by using ``SklearnInstrumentor``.
|
||||||
|
|
||||||
|
.. sklearn: https://github.com/scikit-learn/scikit-learn
|
||||||
|
|
||||||
|
Usage
|
||||||
|
-----
|
||||||
|
|
||||||
|
Package instrumentation example:
|
||||||
|
|
||||||
|
.. code-block:: python
|
||||||
|
|
||||||
|
from opentelemetry.instrumentation.sklearn import SklearnInstrumentor
|
||||||
|
|
||||||
|
# instrument the sklearn library
|
||||||
|
SklearnInstrumentor().instrument()
|
||||||
|
|
||||||
|
# instrument sklearn and other libraries
|
||||||
|
SklearnInstrumentor(
|
||||||
|
packages=["sklearn", "lightgbm", "xgboost"]
|
||||||
|
).instrument()
|
||||||
|
|
||||||
|
|
||||||
|
Model intrumentation example:
|
||||||
|
|
||||||
|
.. code-block:: python
|
||||||
|
|
||||||
|
from opentelemetry.instrumentation.sklearn import SklearnInstrumentor
|
||||||
|
from sklearn.datasets import load_iris
|
||||||
|
from sklearn.ensemble import RandomForestClassifier
|
||||||
|
from sklearn.model_selection import train_test_split
|
||||||
|
from sklearn.pipeline import Pipeline
|
||||||
|
|
||||||
|
X, y = load_iris(return_X_y=True)
|
||||||
|
X_train, X_test, y_train, y_test = train_test_split(X, y)
|
||||||
|
|
||||||
|
model = Pipeline(
|
||||||
|
[
|
||||||
|
("class", RandomForestClassifier(n_estimators=10)),
|
||||||
|
]
|
||||||
|
)
|
||||||
|
|
||||||
|
model.fit(X_train, y_train)
|
||||||
|
|
||||||
|
SklearnInstrumentor().instrument_estimator(model)
|
||||||
|
|
||||||
|
"""
|
||||||
|
import logging
|
||||||
|
import os
|
||||||
|
from functools import wraps
|
||||||
|
from importlib import import_module
|
||||||
|
from inspect import isclass
|
||||||
|
from pkgutil import iter_modules
|
||||||
|
from typing import Callable, Dict, List, MutableMapping, Sequence, Type, Union
|
||||||
|
|
||||||
|
from sklearn.base import BaseEstimator
|
||||||
|
from sklearn.pipeline import FeatureUnion, Pipeline
|
||||||
|
from sklearn.tree import BaseDecisionTree
|
||||||
|
from sklearn.utils.metaestimators import _IffHasAttrDescriptor
|
||||||
|
|
||||||
|
from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
|
||||||
|
from opentelemetry.instrumentation.sklearn.version import __version__
|
||||||
|
from opentelemetry.trace import get_tracer
|
||||||
|
from opentelemetry.util.types import Attributes
|
||||||
|
|
||||||
|
logger = logging.getLogger(__name__)
|
||||||
|
|
||||||
|
|
||||||
|
def implement_span_estimator(
|
||||||
|
func: Callable,
|
||||||
|
estimator: Union[BaseEstimator, Type[BaseEstimator]],
|
||||||
|
attributes: Attributes = None,
|
||||||
|
):
|
||||||
|
"""Wrap the method call with a span.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
func: A callable to be wrapped in a span
|
||||||
|
estimator: An instance or class of an estimator
|
||||||
|
attributes: Attributes to apply to the span
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
The passed function wrapped in a span.
|
||||||
|
"""
|
||||||
|
if isclass(estimator):
|
||||||
|
name = estimator.__name__
|
||||||
|
else:
|
||||||
|
name = estimator.__class__.__name__
|
||||||
|
logger.debug("Instrumenting: %s.%s", name, func.__name__)
|
||||||
|
attributes = attributes or {}
|
||||||
|
name = "{cls}.{func}".format(cls=name, func=func.__name__)
|
||||||
|
return implement_span_function(func, name, attributes)
|
||||||
|
|
||||||
|
|
||||||
|
def implement_span_function(func: Callable, name: str, attributes: Attributes):
|
||||||
|
"""Wrap the function with a span.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
func: A callable to be wrapped in a span
|
||||||
|
name: The name of the span
|
||||||
|
attributes: Attributes to apply to the span
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
The passed function wrapped in a span.
|
||||||
|
"""
|
||||||
|
|
||||||
|
@wraps(func)
|
||||||
|
def wrapper(*args, **kwargs):
|
||||||
|
with get_tracer(__name__, __version__).start_as_current_span(
|
||||||
|
name=name
|
||||||
|
) as span:
|
||||||
|
if span.is_recording():
|
||||||
|
for key, val in attributes.items():
|
||||||
|
span.set_attribute(key, val)
|
||||||
|
return func(*args, **kwargs)
|
||||||
|
|
||||||
|
return wrapper
|
||||||
|
|
||||||
|
|
||||||
|
def implement_span_delegator(
|
||||||
|
obj: _IffHasAttrDescriptor, attributes: Attributes = None
|
||||||
|
):
|
||||||
|
"""Wrap the descriptor's fn with a span.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
obj: An instance of _IffHasAttrDescriptor
|
||||||
|
attributes: Attributes to apply to the span
|
||||||
|
"""
|
||||||
|
# Don't instrument inherited delegators
|
||||||
|
if hasattr(obj, "_otel_original_fn"):
|
||||||
|
logger.debug("Already instrumented: %s", obj.fn.__qualname__)
|
||||||
|
return
|
||||||
|
logger.debug("Instrumenting: %s", obj.fn.__qualname__)
|
||||||
|
attributes = attributes or {}
|
||||||
|
setattr(obj, "_otel_original_fn", getattr(obj, "fn"))
|
||||||
|
setattr(
|
||||||
|
obj,
|
||||||
|
"fn",
|
||||||
|
implement_span_function(obj.fn, obj.fn.__qualname__, attributes),
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
def get_delegator(
|
||||||
|
estimator: Type[BaseEstimator], method_name: str
|
||||||
|
) -> Union[_IffHasAttrDescriptor, None]:
|
||||||
|
"""Get the delegator from a class method or None.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator: A class derived from ``sklearn``'s ``BaseEstimator``.
|
||||||
|
method_name (str): The method name of the estimator on which to
|
||||||
|
check for delegation.
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
The delegator, if one exists, otherwise None.
|
||||||
|
"""
|
||||||
|
class_attr = getattr(estimator, method_name)
|
||||||
|
if getattr(class_attr, "__closure__", None) is not None:
|
||||||
|
for cell in class_attr.__closure__:
|
||||||
|
if isinstance(cell.cell_contents, _IffHasAttrDescriptor):
|
||||||
|
return cell.cell_contents
|
||||||
|
return None
|
||||||
|
|
||||||
|
|
||||||
|
def get_base_estimators(packages: List[str]) -> Dict[str, Type[BaseEstimator]]:
|
||||||
|
"""Walk package hierarchies to get BaseEstimator-derived classes.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
packages (list(str)): A list of package names to instrument.
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
A dictionary of qualnames and classes inheriting from
|
||||||
|
``BaseEstimator``.
|
||||||
|
"""
|
||||||
|
klasses = dict()
|
||||||
|
for package_name in packages:
|
||||||
|
lib = import_module(package_name)
|
||||||
|
package_dir = os.path.dirname(lib.__file__)
|
||||||
|
for (_, module_name, _) in iter_modules([package_dir]):
|
||||||
|
# import the module and iterate through its attributes
|
||||||
|
try:
|
||||||
|
module = import_module(package_name + "." + module_name)
|
||||||
|
except ImportError:
|
||||||
|
logger.warning(
|
||||||
|
"Unable to import %s.%s", package_name, module_name
|
||||||
|
)
|
||||||
|
continue
|
||||||
|
for attribute_name in dir(module):
|
||||||
|
attrib = getattr(module, attribute_name)
|
||||||
|
if isclass(attrib) and issubclass(attrib, BaseEstimator):
|
||||||
|
klasses[
|
||||||
|
".".join([package_name, module_name, attribute_name])
|
||||||
|
] = attrib
|
||||||
|
return klasses
|
||||||
|
|
||||||
|
|
||||||
|
# Methods on which spans should be applied.
|
||||||
|
DEFAULT_METHODS = [
|
||||||
|
"fit",
|
||||||
|
"transform",
|
||||||
|
"predict",
|
||||||
|
"predict_proba",
|
||||||
|
"_fit",
|
||||||
|
"_transform",
|
||||||
|
"_predict",
|
||||||
|
"_predict_proba",
|
||||||
|
]
|
||||||
|
|
||||||
|
# Classes and their attributes which contain a list of tupled estimators
|
||||||
|
# through which we should walk recursively for estimators.
|
||||||
|
DEFAULT_NAMEDTUPLE_ATTRIBS = {
|
||||||
|
Pipeline: ["steps"],
|
||||||
|
FeatureUnion: ["transformer_list"],
|
||||||
|
}
|
||||||
|
|
||||||
|
# Classes and their attributes which contain an estimator or sequence of
|
||||||
|
# estimators through which we should walk recursively for estimators.
|
||||||
|
DEFAULT_ATTRIBS = {}
|
||||||
|
|
||||||
|
# Classes (including children) explicitly excluded from autoinstrumentation
|
||||||
|
DEFAULT_EXCLUDE_CLASSES = [BaseDecisionTree]
|
||||||
|
|
||||||
|
# Default packages for autoinstrumentation
|
||||||
|
DEFAULT_PACKAGES = ["sklearn"]
|
||||||
|
|
||||||
|
|
||||||
|
class SklearnInstrumentor(BaseInstrumentor):
|
||||||
|
"""Instrument a fitted sklearn model with opentelemetry spans.
|
||||||
|
|
||||||
|
Instrument methods of ``BaseEstimator``-derived components in a sklearn
|
||||||
|
model. The assumption is that a machine learning model ``Pipeline`` (or
|
||||||
|
class descendent) is being instrumented with opentelemetry. Within a
|
||||||
|
``Pipeline`` is some hierarchy of estimators and transformers.
|
||||||
|
|
||||||
|
The ``instrument_estimator`` method walks this hierarchy of estimators,
|
||||||
|
implementing each of the defined methods with its own span.
|
||||||
|
|
||||||
|
Certain estimators in the sklearn ecosystem contain other estimators as
|
||||||
|
instance attributes. Support for walking this embedded sub-hierarchy is
|
||||||
|
supported with ``recurse_attribs``. This argument is a dictionary
|
||||||
|
with classes as keys, and a list of attributes representing embedded
|
||||||
|
estimators as values. By default, ``recurse_attribs`` is empty.
|
||||||
|
|
||||||
|
Similar to Pipelines, there are also estimators which have class attributes
|
||||||
|
as a list of 2-tuples; for instance, the ``FeatureUnion`` and its attribute
|
||||||
|
``transformer_list``. Instrumenting estimators like this is also
|
||||||
|
supported through the ``recurse_namedtuple_attribs`` argument. This
|
||||||
|
argument is a dictionary with classes as keys, and a list of attribute
|
||||||
|
names representing the namedtuple list(s). By default, the
|
||||||
|
``recurse_namedtuple_attribs`` dictionary supports
|
||||||
|
``Pipeline`` with ``steps``, and ``FeatureUnion`` with
|
||||||
|
``transformer_list``.
|
||||||
|
|
||||||
|
Note that spans will not be generated for any child transformer whose
|
||||||
|
parent transformer has ``n_jobs`` parameter set to anything besides
|
||||||
|
``None`` or ``1``.
|
||||||
|
|
||||||
|
Package instrumentation example:
|
||||||
|
|
||||||
|
.. code-block:: python
|
||||||
|
|
||||||
|
from opentelemetry.instrumentation.sklearn import SklearnInstrumentor
|
||||||
|
|
||||||
|
# instrument the sklearn library
|
||||||
|
SklearnInstrumentor().instrument()
|
||||||
|
|
||||||
|
# instrument several sklearn-compatible libraries
|
||||||
|
packages = ["sklearn", "lightgbm", "xgboost"]
|
||||||
|
SklearnInstrumentor(packages=packages).instrument()
|
||||||
|
|
||||||
|
|
||||||
|
Model intrumentation example:
|
||||||
|
|
||||||
|
.. code-block:: python
|
||||||
|
|
||||||
|
from opentelemetry.instrumentation.sklearn import SklearnInstrumentor
|
||||||
|
from sklearn.datasets import load_iris
|
||||||
|
from sklearn.ensemble import RandomForestClassifier
|
||||||
|
from sklearn.model_selection import train_test_split
|
||||||
|
from sklearn.pipeline import Pipeline
|
||||||
|
|
||||||
|
X, y = load_iris(return_X_y=True)
|
||||||
|
X_train, X_test, y_train, y_test = train_test_split(X, y)
|
||||||
|
|
||||||
|
model = Pipeline(
|
||||||
|
[
|
||||||
|
("class", RandomForestClassifier(n_estimators=10)),
|
||||||
|
]
|
||||||
|
)
|
||||||
|
|
||||||
|
model.fit(X_train, y_train)
|
||||||
|
|
||||||
|
SklearnInstrumentor().instrument_estimator(model)
|
||||||
|
|
||||||
|
Args:
|
||||||
|
methods (list): A list of method names on which to instrument a span.
|
||||||
|
This list of methods will be checked on all estimators in the model
|
||||||
|
hierarchy. Used in package and model instrumentation
|
||||||
|
recurse_attribs (dict): A dictionary of ``BaseEstimator``-derived
|
||||||
|
sklearn classes as keys, with values being a list of attributes. Each
|
||||||
|
attribute represents either an estimator or list of estimators on
|
||||||
|
which to also implement spans. An example is
|
||||||
|
``RandomForestClassifier`` and its attribute ``estimators_``. Used
|
||||||
|
in model instrumentation only.
|
||||||
|
recurse_namedtuple_attribs (dict): A dictionary of ``BaseEstimator``-
|
||||||
|
derived sklearn types as keys, with values being a list of
|
||||||
|
attribute names. Each attribute represents a list of 2-tuples in
|
||||||
|
which the first element is the estimator name, and the second
|
||||||
|
element is the estimator. Defaults include sklearn's ``Pipeline``
|
||||||
|
and its attribute ``steps``, and the ``FeatureUnion`` and its
|
||||||
|
attribute ``transformer_list``. Used in model instrumentation only.
|
||||||
|
packages: A list of sklearn-compatible packages to
|
||||||
|
instrument. Used with package instrumentation only.
|
||||||
|
exclude_classes: A list of classes to exclude from instrumentation.
|
||||||
|
Child classes are also excluded. Default is sklearn's
|
||||||
|
``[BaseDecisionTree]``.
|
||||||
|
"""
|
||||||
|
|
||||||
|
def __new__(cls, *args, **kwargs):
|
||||||
|
"""Override new.
|
||||||
|
|
||||||
|
The base class' new method passes args and kwargs. We override because
|
||||||
|
we init the class with configuration and Python raises TypeError when
|
||||||
|
additional arguments are passed to the object.__new__() method.
|
||||||
|
"""
|
||||||
|
if cls._instance is None:
|
||||||
|
cls._instance = object.__new__(cls)
|
||||||
|
|
||||||
|
return cls._instance
|
||||||
|
|
||||||
|
def __init__(
|
||||||
|
self,
|
||||||
|
methods: List[str] = None,
|
||||||
|
recurse_attribs: Dict[Type[BaseEstimator], List[str]] = None,
|
||||||
|
recurse_namedtuple_attribs: Dict[
|
||||||
|
Type[BaseEstimator], List[str]
|
||||||
|
] = None,
|
||||||
|
packages: List[str] = None,
|
||||||
|
exclude_classes: List[Type] = None,
|
||||||
|
):
|
||||||
|
self.methods = methods or DEFAULT_METHODS
|
||||||
|
self.recurse_attribs = recurse_attribs or DEFAULT_ATTRIBS
|
||||||
|
self.recurse_namedtuple_attribs = (
|
||||||
|
recurse_namedtuple_attribs or DEFAULT_NAMEDTUPLE_ATTRIBS
|
||||||
|
)
|
||||||
|
self.packages = packages or DEFAULT_PACKAGES
|
||||||
|
if exclude_classes is None:
|
||||||
|
self.exclude_classes = tuple(DEFAULT_EXCLUDE_CLASSES)
|
||||||
|
else:
|
||||||
|
self.exclude_classes = tuple(exclude_classes)
|
||||||
|
|
||||||
|
def _instrument(self, **kwargs):
|
||||||
|
"""Instrument the library, and any additional specified on init."""
|
||||||
|
klasses = get_base_estimators(packages=self.packages)
|
||||||
|
attributes = kwargs.get("attributes")
|
||||||
|
for _, klass in klasses.items():
|
||||||
|
if issubclass(klass, self.exclude_classes):
|
||||||
|
logger.debug("Not instrumenting (excluded): %s", str(klass))
|
||||||
|
else:
|
||||||
|
logger.debug("Instrumenting: %s", str(klass))
|
||||||
|
for method_name in self.methods:
|
||||||
|
if hasattr(klass, method_name):
|
||||||
|
self._instrument_class_method(
|
||||||
|
estimator=klass,
|
||||||
|
method_name=method_name,
|
||||||
|
attributes=attributes,
|
||||||
|
)
|
||||||
|
|
||||||
|
def _uninstrument(self, **kwargs):
|
||||||
|
"""Uninstrument the library"""
|
||||||
|
klasses = get_base_estimators(packages=self.packages)
|
||||||
|
for _, klass in klasses.items():
|
||||||
|
logger.debug("Uninstrumenting: %s", str(klass))
|
||||||
|
for method_name in self.methods:
|
||||||
|
if hasattr(klass, method_name):
|
||||||
|
self._uninstrument_class_method(
|
||||||
|
estimator=klass, method_name=method_name
|
||||||
|
)
|
||||||
|
|
||||||
|
def instrument_estimator(
|
||||||
|
self, estimator: BaseEstimator, attributes: Attributes = None
|
||||||
|
):
|
||||||
|
"""Instrument a fitted estimator and its hierarchy where configured.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (sklearn.base.BaseEstimator): A fitted ``sklearn``
|
||||||
|
estimator, typically a ``Pipeline`` instance.
|
||||||
|
attributes (dict): Attributes to attach to the spans.
|
||||||
|
"""
|
||||||
|
if isinstance(estimator, self.exclude_classes):
|
||||||
|
logger.debug(
|
||||||
|
"Not instrumenting (excluded): %s",
|
||||||
|
estimator.__class__.__name__,
|
||||||
|
)
|
||||||
|
return
|
||||||
|
|
||||||
|
if isinstance(
|
||||||
|
estimator, tuple(self.recurse_namedtuple_attribs.keys())
|
||||||
|
):
|
||||||
|
self._instrument_estimator_namedtuple(
|
||||||
|
estimator=estimator, attributes=attributes
|
||||||
|
)
|
||||||
|
|
||||||
|
if isinstance(estimator, tuple(self.recurse_attribs.keys())):
|
||||||
|
self._instrument_estimator_attribute(
|
||||||
|
estimator=estimator, attributes=attributes
|
||||||
|
)
|
||||||
|
|
||||||
|
for method_name in self.methods:
|
||||||
|
if hasattr(estimator, method_name):
|
||||||
|
self._instrument_instance_method(
|
||||||
|
estimator=estimator,
|
||||||
|
method_name=method_name,
|
||||||
|
attributes=attributes,
|
||||||
|
)
|
||||||
|
|
||||||
|
def uninstrument_estimator(self, estimator: BaseEstimator):
|
||||||
|
"""Uninstrument a fitted estimator and its hierarchy where configured.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (sklearn.base.BaseEstimator): A fitted ``sklearn``
|
||||||
|
estimator, typically a ``Pipeline`` instance.
|
||||||
|
"""
|
||||||
|
if isinstance(estimator, self.exclude_classes):
|
||||||
|
logger.debug(
|
||||||
|
"Not uninstrumenting (excluded): %s",
|
||||||
|
estimator.__class__.__name__,
|
||||||
|
)
|
||||||
|
return
|
||||||
|
|
||||||
|
if isinstance(
|
||||||
|
estimator, tuple(self.recurse_namedtuple_attribs.keys())
|
||||||
|
):
|
||||||
|
self._uninstrument_estimator_namedtuple(estimator=estimator)
|
||||||
|
|
||||||
|
if isinstance(estimator, tuple(self.recurse_attribs.keys())):
|
||||||
|
self._uninstrument_estimator_attribute(estimator=estimator)
|
||||||
|
|
||||||
|
for method_name in self.methods:
|
||||||
|
if hasattr(estimator, method_name):
|
||||||
|
self._uninstrument_instance_method(
|
||||||
|
estimator=estimator, method_name=method_name
|
||||||
|
)
|
||||||
|
|
||||||
|
def _check_instrumented(
|
||||||
|
self,
|
||||||
|
estimator: Union[BaseEstimator, Type[BaseEstimator]],
|
||||||
|
method_name: str,
|
||||||
|
) -> bool:
|
||||||
|
"""Check an estimator-method is instrumented.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A class or instance of an ``sklearn``
|
||||||
|
estimator.
|
||||||
|
method_name (str): The method name of the estimator on which to
|
||||||
|
check for instrumentation.
|
||||||
|
"""
|
||||||
|
orig_method_name = "_otel_original_" + method_name
|
||||||
|
has_original = hasattr(estimator, orig_method_name)
|
||||||
|
orig_class, orig_method = getattr(
|
||||||
|
estimator, orig_method_name, (None, None)
|
||||||
|
)
|
||||||
|
same_class = orig_class == estimator
|
||||||
|
if has_original and same_class:
|
||||||
|
class_method = self._unwrap_function(
|
||||||
|
getattr(estimator, method_name)
|
||||||
|
)
|
||||||
|
# if they match then the subclass doesn't override
|
||||||
|
# if they don't then the overridden method needs instrumentation
|
||||||
|
if class_method.__name__ == orig_method.__name__:
|
||||||
|
return True
|
||||||
|
return False
|
||||||
|
|
||||||
|
def _uninstrument_class_method(
|
||||||
|
self, estimator: Type[BaseEstimator], method_name: str
|
||||||
|
):
|
||||||
|
"""Uninstrument a class method.
|
||||||
|
|
||||||
|
Replaces the patched method with the original, and deletes the
|
||||||
|
attribute which stored the original method.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A class or instance of an ``sklearn``
|
||||||
|
estimator.
|
||||||
|
method_name (str): The method name of the estimator on which to
|
||||||
|
apply a span.
|
||||||
|
"""
|
||||||
|
orig_method_name = "_otel_original_" + method_name
|
||||||
|
if isclass(estimator):
|
||||||
|
qualname = estimator.__qualname__
|
||||||
|
else:
|
||||||
|
qualname = estimator.__class__.__qualname__
|
||||||
|
delegator = get_delegator(estimator, method_name)
|
||||||
|
if self._check_instrumented(estimator, method_name):
|
||||||
|
logger.debug(
|
||||||
|
"Uninstrumenting: %s.%s", qualname, method_name,
|
||||||
|
)
|
||||||
|
_, orig_method = getattr(estimator, orig_method_name)
|
||||||
|
setattr(
|
||||||
|
estimator, method_name, orig_method,
|
||||||
|
)
|
||||||
|
delattr(estimator, orig_method_name)
|
||||||
|
elif delegator is not None:
|
||||||
|
if not hasattr(delegator, "_otel_original_fn"):
|
||||||
|
logger.debug(
|
||||||
|
"Already uninstrumented: %s.%s", qualname, method_name,
|
||||||
|
)
|
||||||
|
return
|
||||||
|
setattr(
|
||||||
|
delegator, "fn", getattr(delegator, "_otel_original_fn"),
|
||||||
|
)
|
||||||
|
delattr(delegator, "_otel_original_fn")
|
||||||
|
else:
|
||||||
|
logger.debug(
|
||||||
|
"Already uninstrumented: %s.%s", qualname, method_name,
|
||||||
|
)
|
||||||
|
|
||||||
|
def _uninstrument_instance_method(
|
||||||
|
self, estimator: BaseEstimator, method_name: str
|
||||||
|
):
|
||||||
|
"""Uninstrument an instance method.
|
||||||
|
|
||||||
|
Replaces the patched method with the original, and deletes the
|
||||||
|
attribute which stored the original method.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A class or instance of an ``sklearn``
|
||||||
|
estimator.
|
||||||
|
method_name (str): The method name of the estimator on which to
|
||||||
|
apply a span.
|
||||||
|
"""
|
||||||
|
orig_method_name = "_otel_original_" + method_name
|
||||||
|
if isclass(estimator):
|
||||||
|
qualname = estimator.__qualname__
|
||||||
|
else:
|
||||||
|
qualname = estimator.__class__.__qualname__
|
||||||
|
if self._check_instrumented(estimator, method_name):
|
||||||
|
logger.debug(
|
||||||
|
"Uninstrumenting: %s.%s", qualname, method_name,
|
||||||
|
)
|
||||||
|
_, orig_method = getattr(estimator, orig_method_name)
|
||||||
|
setattr(
|
||||||
|
estimator, method_name, orig_method,
|
||||||
|
)
|
||||||
|
delattr(estimator, orig_method_name)
|
||||||
|
else:
|
||||||
|
logger.debug(
|
||||||
|
"Already uninstrumented: %s.%s", qualname, method_name,
|
||||||
|
)
|
||||||
|
|
||||||
|
def _instrument_class_method(
|
||||||
|
self,
|
||||||
|
estimator: Type[BaseEstimator],
|
||||||
|
method_name: str,
|
||||||
|
attributes: Attributes = None,
|
||||||
|
):
|
||||||
|
"""Instrument an estimator method with a span.
|
||||||
|
|
||||||
|
When instrumenting we attach a tuple of (Class, method) to the
|
||||||
|
attribute ``_otel_original_<method_name>`` for each method. This allows
|
||||||
|
us to replace the patched with the original in uninstrumentation, but
|
||||||
|
also allows proper instrumentation of child classes without
|
||||||
|
instrumenting inherited methods twice.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A ``BaseEstimator``-derived
|
||||||
|
class
|
||||||
|
method_name (str): The method name of the estimator on which to
|
||||||
|
apply a span.
|
||||||
|
attributes (dict): Attributes to attach to the spans.
|
||||||
|
"""
|
||||||
|
if self._check_instrumented(estimator, method_name):
|
||||||
|
logger.debug(
|
||||||
|
"Already instrumented: %s.%s",
|
||||||
|
estimator.__qualname__,
|
||||||
|
method_name,
|
||||||
|
)
|
||||||
|
return
|
||||||
|
class_attr = getattr(estimator, method_name)
|
||||||
|
delegator = get_delegator(estimator, method_name)
|
||||||
|
if isinstance(class_attr, property):
|
||||||
|
logger.debug(
|
||||||
|
"Not instrumenting found property: %s.%s",
|
||||||
|
estimator.__qualname__,
|
||||||
|
method_name,
|
||||||
|
)
|
||||||
|
elif delegator is not None:
|
||||||
|
implement_span_delegator(delegator)
|
||||||
|
else:
|
||||||
|
setattr(
|
||||||
|
estimator,
|
||||||
|
"_otel_original_" + method_name,
|
||||||
|
(estimator, class_attr),
|
||||||
|
)
|
||||||
|
setattr(
|
||||||
|
estimator,
|
||||||
|
method_name,
|
||||||
|
implement_span_estimator(class_attr, estimator, attributes),
|
||||||
|
)
|
||||||
|
|
||||||
|
def _unwrap_function(self, function):
|
||||||
|
"""Fetch the function underlying any decorators"""
|
||||||
|
if hasattr(function, "__wrapped__"):
|
||||||
|
return self._unwrap_function(function.__wrapped__)
|
||||||
|
return function
|
||||||
|
|
||||||
|
def _instrument_instance_method(
|
||||||
|
self,
|
||||||
|
estimator: BaseEstimator,
|
||||||
|
method_name: str,
|
||||||
|
attributes: Attributes = None,
|
||||||
|
):
|
||||||
|
"""Instrument an estimator instance method with a span.
|
||||||
|
|
||||||
|
When instrumenting we attach a tuple of (Class, method) to the
|
||||||
|
attribute ``_otel_original_<method_name>`` for each method. This allows
|
||||||
|
us to replace the patched with the original in unstrumentation.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A fitted ``sklearn`` estimator.
|
||||||
|
method_name (str): The method name of the estimator on which to
|
||||||
|
apply a span.
|
||||||
|
attributes (dict): Attributes to attach to the spans.
|
||||||
|
"""
|
||||||
|
if self._check_instrumented(estimator, method_name):
|
||||||
|
logger.debug(
|
||||||
|
"Already instrumented: %s.%s",
|
||||||
|
estimator.__class__.__qualname__,
|
||||||
|
method_name,
|
||||||
|
)
|
||||||
|
return
|
||||||
|
|
||||||
|
class_attr = getattr(type(estimator), method_name, None)
|
||||||
|
if isinstance(class_attr, property):
|
||||||
|
logger.debug(
|
||||||
|
"Not instrumenting found property: %s.%s",
|
||||||
|
estimator.__class__.__qualname__,
|
||||||
|
method_name,
|
||||||
|
)
|
||||||
|
else:
|
||||||
|
method = getattr(estimator, method_name)
|
||||||
|
setattr(
|
||||||
|
estimator, "_otel_original_" + method_name, (estimator, method)
|
||||||
|
)
|
||||||
|
setattr(
|
||||||
|
estimator,
|
||||||
|
method_name,
|
||||||
|
implement_span_estimator(method, estimator, attributes),
|
||||||
|
)
|
||||||
|
|
||||||
|
def _instrument_estimator_attribute(
|
||||||
|
self, estimator: BaseEstimator, attributes: Attributes = None
|
||||||
|
):
|
||||||
|
"""Instrument instance attributes which also contain estimators.
|
||||||
|
|
||||||
|
Handle instance attributes which are also estimators, are a list
|
||||||
|
(Sequence) of estimators, or are mappings (dictionary) in which
|
||||||
|
the values are estimators.
|
||||||
|
|
||||||
|
Examples include ``RandomForestClassifier`` and
|
||||||
|
``MultiOutputRegressor`` instances which have attributes
|
||||||
|
``estimators_`` attributes.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A fitted ``sklearn`` estimator, with an
|
||||||
|
attribute which also contains an estimator or collection of
|
||||||
|
estimators.
|
||||||
|
attributes (dict): Attributes to attach to the spans.
|
||||||
|
"""
|
||||||
|
attribs = self.recurse_attribs.get(estimator.__class__, [])
|
||||||
|
for attrib in attribs:
|
||||||
|
attrib_value = getattr(estimator, attrib)
|
||||||
|
if isinstance(attrib_value, Sequence):
|
||||||
|
for value in attrib_value:
|
||||||
|
self.instrument_estimator(
|
||||||
|
estimator=value, attributes=attributes
|
||||||
|
)
|
||||||
|
elif isinstance(attrib_value, MutableMapping):
|
||||||
|
for value in attrib_value.values():
|
||||||
|
self.instrument_estimator(
|
||||||
|
estimator=value, attributes=attributes
|
||||||
|
)
|
||||||
|
else:
|
||||||
|
self.instrument_estimator(
|
||||||
|
estimator=attrib_value, attributes=attributes
|
||||||
|
)
|
||||||
|
|
||||||
|
def _instrument_estimator_namedtuple(
|
||||||
|
self, estimator: BaseEstimator, attributes: Attributes = None
|
||||||
|
):
|
||||||
|
"""Instrument attributes with (name, estimator) tupled components.
|
||||||
|
|
||||||
|
Examples include Pipeline and FeatureUnion instances which
|
||||||
|
have attributes steps and transformer_list, respectively.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator: A fitted sklearn estimator, with an attribute which also
|
||||||
|
contains an estimator or collection of estimators.
|
||||||
|
attributes (dict): Attributes to attach to the spans.
|
||||||
|
"""
|
||||||
|
attribs = self.recurse_namedtuple_attribs.get(estimator.__class__, [])
|
||||||
|
for attrib in attribs:
|
||||||
|
for _, est in getattr(estimator, attrib):
|
||||||
|
self.instrument_estimator(estimator=est, attributes=attributes)
|
||||||
|
|
||||||
|
def _uninstrument_estimator_attribute(self, estimator: BaseEstimator):
|
||||||
|
"""Uninstrument instance attributes which also contain estimators.
|
||||||
|
|
||||||
|
Handle instance attributes which are also estimators, are a list
|
||||||
|
(Sequence) of estimators, or are mappings (dictionary) in which
|
||||||
|
the values are estimators.
|
||||||
|
|
||||||
|
Examples include ``RandomForestClassifier`` and
|
||||||
|
``MultiOutputRegressor`` instances which have attributes
|
||||||
|
``estimators_`` attributes.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator (BaseEstimator): A fitted ``sklearn`` estimator, with an
|
||||||
|
attribute which also contains an estimator or collection of
|
||||||
|
estimators.
|
||||||
|
"""
|
||||||
|
attribs = self.recurse_attribs.get(estimator.__class__, [])
|
||||||
|
for attrib in attribs:
|
||||||
|
attrib_value = getattr(estimator, attrib)
|
||||||
|
if isinstance(attrib_value, Sequence):
|
||||||
|
for value in attrib_value:
|
||||||
|
self.uninstrument_estimator(estimator=value)
|
||||||
|
elif isinstance(attrib_value, MutableMapping):
|
||||||
|
for value in attrib_value.values():
|
||||||
|
self.uninstrument_estimator(estimator=value)
|
||||||
|
else:
|
||||||
|
self.uninstrument_estimator(estimator=attrib_value)
|
||||||
|
|
||||||
|
def _uninstrument_estimator_namedtuple(self, estimator: BaseEstimator):
|
||||||
|
"""Uninstrument attributes with (name, estimator) tupled components.
|
||||||
|
|
||||||
|
Examples include Pipeline and FeatureUnion instances which
|
||||||
|
have attributes steps and transformer_list, respectively.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
estimator: A fitted sklearn estimator, with an attribute which also
|
||||||
|
contains an estimator or collection of estimators.
|
||||||
|
"""
|
||||||
|
attribs = self.recurse_namedtuple_attribs.get(estimator.__class__, [])
|
||||||
|
for attrib in attribs:
|
||||||
|
for _, est in getattr(estimator, attrib):
|
||||||
|
self.uninstrument_estimator(estimator=est)
|
||||||
@ -0,0 +1,15 @@
|
|||||||
|
# Copyright 2020, OpenTelemetry Authors
|
||||||
|
#
|
||||||
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
# you may not use this file except in compliance with the License.
|
||||||
|
# You may obtain a copy of the License at
|
||||||
|
#
|
||||||
|
# http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
#
|
||||||
|
# Unless required by applicable law or agreed to in writing, software
|
||||||
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
# See the License for the specific language governing permissions and
|
||||||
|
# limitations under the License.
|
||||||
|
|
||||||
|
__version__ = "0.16.dev0"
|
||||||
@ -0,0 +1,54 @@
|
|||||||
|
# Copyright 2020, OpenTelemetry Authors
|
||||||
|
#
|
||||||
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
# you may not use this file except in compliance with the License.
|
||||||
|
# You may obtain a copy of the License at
|
||||||
|
#
|
||||||
|
# http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
#
|
||||||
|
# Unless required by applicable law or agreed to in writing, software
|
||||||
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
# See the License for the specific language governing permissions and
|
||||||
|
# limitations under the License.
|
||||||
|
|
||||||
|
import numpy as np
|
||||||
|
from sklearn.datasets import load_iris
|
||||||
|
from sklearn.decomposition import PCA, TruncatedSVD
|
||||||
|
from sklearn.ensemble import RandomForestClassifier
|
||||||
|
from sklearn.model_selection import train_test_split
|
||||||
|
from sklearn.pipeline import FeatureUnion, Pipeline
|
||||||
|
from sklearn.preprocessing import Normalizer, StandardScaler
|
||||||
|
|
||||||
|
X, y = load_iris(return_X_y=True)
|
||||||
|
X_train, X_test, y_train, y_test = train_test_split(X, y)
|
||||||
|
|
||||||
|
|
||||||
|
def pipeline():
|
||||||
|
"""A dummy model that has a bunch of components that we can test."""
|
||||||
|
model = Pipeline(
|
||||||
|
[
|
||||||
|
("scaler", StandardScaler()),
|
||||||
|
("normal", Normalizer()),
|
||||||
|
(
|
||||||
|
"union",
|
||||||
|
FeatureUnion(
|
||||||
|
[
|
||||||
|
("pca", PCA(n_components=1)),
|
||||||
|
("svd", TruncatedSVD(n_components=2)),
|
||||||
|
],
|
||||||
|
n_jobs=1, # parallelized components won't generate spans
|
||||||
|
),
|
||||||
|
),
|
||||||
|
("class", RandomForestClassifier(n_estimators=10)),
|
||||||
|
]
|
||||||
|
)
|
||||||
|
model.fit(X_train, y_train)
|
||||||
|
return model
|
||||||
|
|
||||||
|
|
||||||
|
def random_input():
|
||||||
|
"""A random record from the feature set."""
|
||||||
|
rows = X.shape[0]
|
||||||
|
random_row = np.random.choice(rows, size=1)
|
||||||
|
return X[random_row, :]
|
||||||
@ -0,0 +1,189 @@
|
|||||||
|
# Copyright 2020, OpenTelemetry Authors
|
||||||
|
#
|
||||||
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
|
# you may not use this file except in compliance with the License.
|
||||||
|
# You may obtain a copy of the License at
|
||||||
|
#
|
||||||
|
# http://www.apache.org/licenses/LICENSE-2.0
|
||||||
|
#
|
||||||
|
# Unless required by applicable law or agreed to in writing, software
|
||||||
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||||
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||||
|
# See the License for the specific language governing permissions and
|
||||||
|
# limitations under the License.
|
||||||
|
|
||||||
|
from sklearn.ensemble import RandomForestClassifier
|
||||||
|
|
||||||
|
from opentelemetry.instrumentation.sklearn import (
|
||||||
|
DEFAULT_EXCLUDE_CLASSES,
|
||||||
|
DEFAULT_METHODS,
|
||||||
|
SklearnInstrumentor,
|
||||||
|
get_base_estimators,
|
||||||
|
get_delegator,
|
||||||
|
)
|
||||||
|
from opentelemetry.test.test_base import TestBase
|
||||||
|
from opentelemetry.trace import SpanKind
|
||||||
|
|
||||||
|
from .fixtures import pipeline, random_input
|
||||||
|
|
||||||
|
|
||||||
|
def assert_instrumented(base_estimators):
|
||||||
|
for _, estimator in base_estimators.items():
|
||||||
|
for method_name in DEFAULT_METHODS:
|
||||||
|
original_method_name = "_otel_original_" + method_name
|
||||||
|
if issubclass(estimator, tuple(DEFAULT_EXCLUDE_CLASSES)):
|
||||||
|
assert not hasattr(estimator, original_method_name)
|
||||||
|
continue
|
||||||
|
class_attr = getattr(estimator, method_name, None)
|
||||||
|
if isinstance(class_attr, property):
|
||||||
|
assert not hasattr(estimator, original_method_name)
|
||||||
|
continue
|
||||||
|
delegator = None
|
||||||
|
if hasattr(estimator, method_name):
|
||||||
|
delegator = get_delegator(estimator, method_name)
|
||||||
|
if delegator is not None:
|
||||||
|
assert hasattr(delegator, "_otel_original_fn")
|
||||||
|
elif hasattr(estimator, method_name):
|
||||||
|
assert hasattr(estimator, original_method_name)
|
||||||
|
|
||||||
|
|
||||||
|
def assert_uninstrumented(base_estimators):
|
||||||
|
for _, estimator in base_estimators.items():
|
||||||
|
for method_name in DEFAULT_METHODS:
|
||||||
|
original_method_name = "_otel_original_" + method_name
|
||||||
|
if issubclass(estimator, tuple(DEFAULT_EXCLUDE_CLASSES)):
|
||||||
|
assert not hasattr(estimator, original_method_name)
|
||||||
|
continue
|
||||||
|
class_attr = getattr(estimator, method_name, None)
|
||||||
|
if isinstance(class_attr, property):
|
||||||
|
assert not hasattr(estimator, original_method_name)
|
||||||
|
continue
|
||||||
|
delegator = None
|
||||||
|
if hasattr(estimator, method_name):
|
||||||
|
delegator = get_delegator(estimator, method_name)
|
||||||
|
if delegator is not None:
|
||||||
|
assert not hasattr(delegator, "_otel_original_fn")
|
||||||
|
elif hasattr(estimator, method_name):
|
||||||
|
assert not hasattr(estimator, original_method_name)
|
||||||
|
|
||||||
|
|
||||||
|
class TestSklearn(TestBase):
|
||||||
|
def test_package_instrumentation(self):
|
||||||
|
ski = SklearnInstrumentor()
|
||||||
|
|
||||||
|
base_estimators = get_base_estimators(packages=["sklearn"])
|
||||||
|
|
||||||
|
model = pipeline()
|
||||||
|
|
||||||
|
ski.instrument()
|
||||||
|
assert_instrumented(base_estimators)
|
||||||
|
|
||||||
|
x_test = random_input()
|
||||||
|
|
||||||
|
model.predict(x_test)
|
||||||
|
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
self.assertEqual(len(spans), 8)
|
||||||
|
self.memory_exporter.clear()
|
||||||
|
|
||||||
|
ski.uninstrument()
|
||||||
|
assert_uninstrumented(base_estimators)
|
||||||
|
|
||||||
|
model = pipeline()
|
||||||
|
x_test = random_input()
|
||||||
|
|
||||||
|
model.predict(x_test)
|
||||||
|
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
self.assertEqual(len(spans), 0)
|
||||||
|
|
||||||
|
def test_span_properties(self):
|
||||||
|
"""Test that we get all of the spans we expect."""
|
||||||
|
model = pipeline()
|
||||||
|
ski = SklearnInstrumentor()
|
||||||
|
ski.instrument_estimator(estimator=model)
|
||||||
|
|
||||||
|
x_test = random_input()
|
||||||
|
|
||||||
|
model.predict(x_test)
|
||||||
|
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
self.assertEqual(len(spans), 8)
|
||||||
|
span = spans[0]
|
||||||
|
self.assertEqual(span.name, "StandardScaler.transform")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[-1].context.span_id)
|
||||||
|
span = spans[1]
|
||||||
|
self.assertEqual(span.name, "Normalizer.transform")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[-1].context.span_id)
|
||||||
|
span = spans[2]
|
||||||
|
self.assertEqual(span.name, "PCA.transform")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[4].context.span_id)
|
||||||
|
span = spans[3]
|
||||||
|
self.assertEqual(span.name, "TruncatedSVD.transform")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[4].context.span_id)
|
||||||
|
span = spans[4]
|
||||||
|
self.assertEqual(span.name, "FeatureUnion.transform")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[-1].context.span_id)
|
||||||
|
span = spans[5]
|
||||||
|
self.assertEqual(span.name, "RandomForestClassifier.predict_proba")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[6].context.span_id)
|
||||||
|
span = spans[6]
|
||||||
|
self.assertEqual(span.name, "RandomForestClassifier.predict")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
self.assertEqual(span.parent.span_id, spans[-1].context.span_id)
|
||||||
|
span = spans[7]
|
||||||
|
self.assertEqual(span.name, "Pipeline.predict")
|
||||||
|
self.assertEqual(span.kind, SpanKind.INTERNAL)
|
||||||
|
|
||||||
|
self.memory_exporter.clear()
|
||||||
|
|
||||||
|
# uninstrument
|
||||||
|
ski.uninstrument_estimator(estimator=model)
|
||||||
|
x_test = random_input()
|
||||||
|
model.predict(x_test)
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
self.assertEqual(len(spans), 0)
|
||||||
|
|
||||||
|
def test_attrib_config(self):
|
||||||
|
"""Test that the attribute config makes spans on the decision trees."""
|
||||||
|
model = pipeline()
|
||||||
|
attrib_config = {RandomForestClassifier: ["estimators_"]}
|
||||||
|
ski = SklearnInstrumentor(
|
||||||
|
recurse_attribs=attrib_config,
|
||||||
|
exclude_classes=[], # decision trees excluded by default
|
||||||
|
)
|
||||||
|
ski.instrument_estimator(estimator=model)
|
||||||
|
|
||||||
|
x_test = random_input()
|
||||||
|
model.predict(x_test)
|
||||||
|
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
self.assertEqual(len(spans), 8 + model.steps[-1][-1].n_estimators)
|
||||||
|
|
||||||
|
self.memory_exporter.clear()
|
||||||
|
|
||||||
|
ski.uninstrument_estimator(estimator=model)
|
||||||
|
x_test = random_input()
|
||||||
|
model.predict(x_test)
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
self.assertEqual(len(spans), 0)
|
||||||
|
|
||||||
|
def test_span_attributes(self):
|
||||||
|
model = pipeline()
|
||||||
|
attributes = {"model_name": "random_forest_model"}
|
||||||
|
ski = SklearnInstrumentor()
|
||||||
|
ski.instrument_estimator(estimator=model, attributes=attributes)
|
||||||
|
|
||||||
|
x_test = random_input()
|
||||||
|
|
||||||
|
model.predict(x_test)
|
||||||
|
|
||||||
|
spans = self.memory_exporter.get_finished_spans()
|
||||||
|
for span in spans:
|
||||||
|
assert span.attributes["model_name"] == "random_forest_model"
|
||||||
Reference in New Issue
Block a user