Update benchmark datasets for classification and segmentation

[rajeshgangireddy]

Summary

Checklist

• I have added unit tests to cover my changes.​
• I have added integration tests to cover my changes.​
• I have ran e2e tests and there is no issues.
• I have added the description of my changes into CHANGELOG in my target branch (e.g., CHANGELOG in develop).​
• I have updated the documentation in my target branch accordingly (e.g., documentation in develop).
• I have linked related issues.

License

• I submit my code changes under the same Apache License that covers the project.
  Feel free to contact the maintainers if that's a concern.
• I have updated the license header for each file (see an example below).

# Copyright (C) 2025 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

[Copilot]

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

Comments suppressed due to low confidence (4)

tests/perf_v2/tasks/semantic_segmentation.py:32

• [nitpick] Verify that the dataset name 'tiny_human_railway_animal' aligns with the project's naming conventions for segmentation benchmarks and is descriptive enough for its content.

name="tiny_human_railway_animal",

tests/perf_v2/tasks/classification.py:55

• [nitpick] Ensure that the dataset name 'multiclass_tiny_pneumonia' is consistent with naming patterns used across classification test cases in the project.

name="multiclass_tiny_pneumonia",

tests/perf/test_semantic_segmentation.py:31

• [nitpick] Confirm that the updated dataset name 'tiny_human_railway_animal' is in line with the overall naming scheme for segmentation benchmarks in tests.

name="tiny_human_railway_animal",

tests/perf/test_classification.py:44

• [nitpick] Check that the dataset name 'multiclass_tiny_pneumonia' is descriptive and consistent with similar dataset naming patterns in the classification performance tests.

name="multiclass_tiny_pneumonia",

[kprokofi]

Can we update this only for perf_v2? First version will be deleted, I hope

[rajeshgangireddy]

Can we update this only for perf_v2? First version will be deleted, I hope

Good idea. Done.

[Copilot]

Pull Request Overview

This PR updates the benchmark datasets used for classification and semantic segmentation tasks by replacing the previously loop‐generated dataset configurations with explicit definitions. Key changes include updated dataset names, new dataset paths, and clearer grouping of datasets by size.

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
tests/perf_v2/tasks/semantic_segmentation.py	Replaces loop-generated dataset definitions with explicit entries for tiny, small, medium, and large segmentation datasets.
tests/perf_v2/tasks/classification.py	Updates and expands dataset definitions for multi-class, multi-label, and h-label classification tasks.

Comments suppressed due to low confidence (2)

tests/perf_v2/tasks/semantic_segmentation.py:32

• [nitpick] The dataset name 'tiny_human_railway_animal' may be too generic if multiple tiny datasets are expected. Consider updating the name to better reflect its content or purpose, if applicable.

name="tiny_human_railway_animal",

tests/perf_v2/tasks/classification.py:55

• [nitpick] The dataset name 'multiclass_tiny_pneumonia' should be reviewed for consistency with other naming patterns. Ensure that the name clearly indicates its content and is aligned with the overall dataset naming conventions.

name="multiclass_tiny_pneumonia",