fix: replace deprecated pkg_resources by desaxce · Pull Request #7026 · tensorflow/tensorboard

desaxce · 2025-11-02T20:39:46Z

Motivation for features / changes

Fixes #7003
When I run tensorboard, I get the deprecation warning, which I don't want:

.venv/lib/python3.12/site-packages/tensorboard/default.py:30: U
serWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pk
g_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refr
ain from using this package or pin to Setuptools<81.
  import pkg_resources

Technical description of changes

I replaced all occurrences of pkg_resources by importlib or packaging.version.

Screenshots of UI changes (or N/A)

==> not applicable

Detailed steps to verify changes work correctly (as executed by you)

On both master and replace_pkg_resources branches, I ran bazel run //tensorboard -- --logdir runs with outputs:

Output on master:

(tf) [precision tensorboard] (master)$ bazel run //tensorboard -- --logdir runs
WARNING: Download from https://mirror.bazel.build/repo1.maven.org/maven2/com/google/guava/guava/32.1.1-jre/guava-32.1.1-jre.jar failed: class java.io.FileNotFoundException GET returned 404 Not Found
WARNING: Download from https://mirror.bazel.build/repo1.maven.org/maven2/com/google/errorprone/error_prone_annotations/2.23.0/error_prone_annotations-2.23.0.jar failed: class java.io.FileNotFoundException GET returned 404 Not Found
WARNING: Download from https://mirror.bazel.build/repo1.maven.org/maven2/com/google/guava/failureaccess/1.0.1/failureaccess-1.0.1.jar failed: class java.io.FileNotFoundException GET returned 404 Not Found
INFO: Analyzed target //tensorboard:tensorboard (0 packages loaded, 0 targets configured).
INFO: Found 1 target...
Target //tensorboard:tensorboard up-to-date:
  bazel-bin/tensorboard/tensorboard
INFO: Elapsed time: 0.234s, Critical Path: 0.00s
INFO: 1 process: 1 internal.
INFO: Build completed successfully, 1 total action
INFO: Running command line: bazel-bin/tensorboard/tensorboard --logdir runs
/home/precision/.cache/bazel/_bazel_precision/cad60362e6aad31575bd1c7909bd969b/execroot/org_tensorflow_tensorboard/bazel-out/k8-fastbuild/bin/tensorboard/tensorboard.runfiles/org_tensorflow_tensorboard/tensorboard/default.py:30: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
  import pkg_resources
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1763289588.190337  871824 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
I0000 00:00:1763289588.190679  871824 cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.
I0000 00:00:1763289588.222931  871824 cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
TensorFlow installation not found - running with reduced feature set.
E0000 00:00:1763289588.872919  871824 cuda_executor.cc:1695] INTERNAL: CUDA Runtime error: Failed call to cudaGetRuntimeVersion: Error loading CUDA libraries. GPU will not be used.: Error loading CUDA libraries. GPU will not be used.
W0000 00:00:1763289588.898912  871824 cuda_executor.cc:1802] GPU interconnect information not available: INTERNAL: NVML doesn't support extracting fabric info or NVLink is not used by the device.
W0000 00:00:1763289588.904227  871824 gpu_device.cc:2362] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...

NOTE: Using experimental fast data loading logic. To disable, pass
    "--load_fast=false" and report issues on GitHub. More details:
    https://github.com/tensorflow/tensorboard/issues/4784

Serving TensorBoard on localhost; to expose to the network, use a proxy or pass --bind_all
TensorBoard 2.21.0a0 at http://localhost:6007/ (Press CTRL+C to quit)

Output on replace_pkg_resources (no pkg_resources warning):

(tf) [precision tensorboard] (replace_pkg_resources)$ bazel run //tensorboard -- --logdir runsWARNING: Download from https://mirror.bazel.build/repo1.maven.org/maven2/com/google/guava/guava/32.1.1-jre/guava-32.1.1-jre.jar failed: class java.io.FileNotFoundException GET returned 404 Not Found
WARNING: Download from https://mirror.bazel.build/repo1.maven.org/maven2/com/google/errorprone/error_prone_annotations/2.23.0/error_prone_annotations-2.23.0.jar failed: class java.io.FileNotFoundException GET returned 404 Not Found
WARNING: Download from https://mirror.bazel.build/repo1.maven.org/maven2/com/google/guava/failureaccess/1.0.1/failureaccess-1.0.1.jar failed: class java.io.FileNotFoundException GET returned 404 Not Found
INFO: Analyzed target //tensorboard:tensorboard (2 packages loaded, 1744 targets configured).
INFO: Found 1 target...
Target //tensorboard:tensorboard up-to-date:
  bazel-bin/tensorboard/tensorboard
INFO: Elapsed time: 2.550s, Critical Path: 0.13s
INFO: 1 process: 1 internal.
INFO: Build completed successfully, 1 total action
INFO: Running command line: bazel-bin/tensorboard/tensorboard --logdir runs
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1763289626.381133  872688 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
I0000 00:00:1763289626.381500  872688 cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.
I0000 00:00:1763289626.414343  872688 cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
TensorFlow installation not found - running with reduced feature set.
E0000 00:00:1763289627.080864  872688 cuda_executor.cc:1695] INTERNAL: CUDA Runtime error: Failed call to cudaGetRuntimeVersion: Error loading CUDA libraries. GPU will not be used.: Error loading CUDA libraries. GPU will not be used.
W0000 00:00:1763289627.108661  872688 cuda_executor.cc:1802] GPU interconnect information not available: INTERNAL: NVML doesn't support extracting fabric info or NVLink is not used by the device.
W0000 00:00:1763289627.112731  872688 gpu_device.cc:2362] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...

NOTE: Using experimental fast data loading logic. To disable, pass
    "--load_fast=false" and report issues on GitHub. More details:
    https://github.com/tensorflow/tensorboard/issues/4784

Serving TensorBoard on localhost; to expose to the network, use a proxy or pass --bind_all
TensorBoard 2.21.0a0 at http://localhost:6007/ (Press CTRL+C to quit)

Alternate designs / implementations considered (or N/A)

--> not applicable

google-cla · 2025-11-02T20:39:50Z

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

arcra · 2025-11-14T23:38:11Z

Thank you @desaxce ! Sorry for the delay. I will look into this next week!

arcra · 2025-11-25T18:37:04Z

So sorry for the delay again. Our CI workflows are currently broken because we seem to have reached the limit of the provisioned VM for GH Actions, and it'll take a little while to resolve this. I'll come back to check this when that is resolved.

adamhadani · 2025-12-05T17:41:02Z

+1

Sacode · 2026-02-08T15:53:21Z

The package 'pkg_resources' has now been removed:
File ".venv\Lib\site-packages\tensorboard\default.py", line 30, in
import pkg_resources
ModuleNotFoundError: No module named 'pkg_resources'

python -c "import pkg_resources; print('pkg_resources is available')"
Traceback (most recent call last):
File "", line 1, in
ModuleNotFoundError: No module named 'pkg_resources'

The old version of setuptools:
python -c "import pkg_resources; print('pkg_resources is available')"
:1: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
pkg_resources is available

biggusdonzus · 2026-02-19T08:25:50Z

So sorry for the delay again. Our CI workflows are currently broken because we seem to have reached the limit of the provisioned VM for GH Actions, and it'll take a little while to resolve this. I'll come back to check this when that is resolved.

Hi @arcra , any news on this?

…o address a compatibility issue with Tensorboard. * **`setuptools<82.0.0`**: The requirement for `setuptools` has been changed from `>=68.0` to `<82.0.0`. This upper bound is necessary for Tensorboard, as indicated by the added comment referencing tensorflow/tensorboard#7026.

arcra · 2026-02-27T00:00:09Z

tensorboard/data/BUILD

        ":grpc_provider",
        ":ingester",
        "//tensorboard:expect_grpc_installed",
-        "//tensorboard:expect_pkg_resources_installed",


I think we need an equivalent expect_packaging_installed, for this to work with our internal repo.

arcra · 2026-02-27T00:01:07Z

tensorboard/pip_package/requirements.txt

 grpcio >= 1.48.2
 markdown >= 2.6.8
 numpy >= 1.12.0
 # NOTE: The packaging dependency was introduced in order to compare installed


Could you also remove this comment, please?

arcra · 2026-02-27T00:02:01Z

tensorboard/BUILD

-# This is a dummy rule used as a pkg-resources dependency in open-source.
-# We expect pkg-resources to already be installed on the system, e.g., via
-# `pip install setuptools`.
-py_library(name = "expect_pkg_resources_installed")


Here's where it would need to be updated for the new library instead.

arcra · 2026-02-27T00:05:35Z

tensorboard/version_test.py

+        compliant_version = parse_version("1.0.0")
        try:
-            legacy_version = pkg_resources.parse_version("arbitrary string")
+            legacy_version = parse_version("arbitrary string")


I think the replacement for this function should have a consistent behavior for the non-compliant ones, so this try-except shouldn't be necessary anymore.

arcra · 2026-02-27T00:10:15Z

Hello, our CI workflow should be working now (I think).

I just reviewed #7057 thinking it was this PR, and I just realized it's a separate one recently created with practically the same content.

I'd like to thank you again for your contribution.

There are a couple things that are a bit better handled on that other PR, and I made some comments about those in this PR. I'll probably merge whichever is ready to merge first.

fix: replace deprecated pkg_resources

e736e3c

fix: replace all pkg_resources refs

1d7a674

dxqb mentioned this pull request Feb 10, 2026

Deprecated pkg_resources package #7003

Open

arcra reviewed Feb 27, 2026

View reviewed changes

arcra approved these changes Feb 27, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix: replace deprecated pkg_resources#7026

fix: replace deprecated pkg_resources#7026
desaxce wants to merge 2 commits intotensorflow:masterfrom
desaxce:replace_pkg_resources

desaxce commented Nov 2, 2025 •

edited

Loading

Uh oh!

google-cla bot commented Nov 2, 2025

Uh oh!

arcra commented Nov 14, 2025

Uh oh!

arcra commented Nov 25, 2025

Uh oh!

adamhadani commented Dec 5, 2025

Uh oh!

Sacode commented Feb 8, 2026

Uh oh!

biggusdonzus commented Feb 19, 2026

Uh oh!

arcra Feb 27, 2026

Uh oh!

arcra Feb 27, 2026

Uh oh!

arcra Feb 27, 2026

Uh oh!

arcra Feb 27, 2026

Uh oh!

arcra commented Feb 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

desaxce commented Nov 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivation for features / changes

Technical description of changes

Screenshots of UI changes (or N/A)

Detailed steps to verify changes work correctly (as executed by you)

Alternate designs / implementations considered (or N/A)

Uh oh!

google-cla bot commented Nov 2, 2025

Uh oh!

arcra commented Nov 14, 2025

Uh oh!

arcra commented Nov 25, 2025

Uh oh!

adamhadani commented Dec 5, 2025

Uh oh!

Sacode commented Feb 8, 2026

Uh oh!

biggusdonzus commented Feb 19, 2026

Uh oh!

arcra Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

arcra Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

arcra Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

arcra Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

arcra commented Feb 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

desaxce commented Nov 2, 2025 •

edited

Loading