-
Notifications
You must be signed in to change notification settings - Fork 444
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* Add action benchmark test template * Fix vsp perf test * Fix act perf test * Add action to perf benchmark workflow * Fix raw data path * Fix benchmark options * Fix vsp dataset format
- Loading branch information
1 parent
326fddf
commit c4db3af
Showing
8 changed files
with
292 additions
and
56 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,182 @@ | ||
| # Copyright (C) 2024 Intel Corporation | ||
| # SPDX-License-Identifier: Apache-2.0 | ||
|
|
||
| """OTX action perfomance benchmark tests.""" | ||
|
|
||
| from __future__ import annotations | ||
|
|
||
| from pathlib import Path | ||
|
|
||
| import pytest | ||
|
|
||
| from .benchmark import Benchmark | ||
| from .conftest import PerfTestBase | ||
|
|
||
|
|
||
| class TestPerfActionClassification(PerfTestBase): | ||
| """Benchmark action classification.""" | ||
|
|
||
| MODEL_TEST_CASES = [ # noqa: RUF012 | ||
| Benchmark.Model(task="action/action_classification", name="movinet", category="speed"), | ||
| Benchmark.Model(task="action/action_classification", name="x3d", category="accuracy"), | ||
| ] | ||
|
|
||
| DATASET_TEST_CASES = [ # noqa: RUF012 | ||
| Benchmark.Dataset( | ||
| name="ucf-5percent-small", | ||
| path=Path("action/action_classification/ucf_kinetics_5percent_small"), | ||
| group="small", | ||
| num_repeat=5, | ||
| extra_overrides={ | ||
| "train": { | ||
| "max_epochs": "10", | ||
| "deterministic": "True", | ||
| }, | ||
| }, | ||
| ), | ||
| Benchmark.Dataset( | ||
| name="ucf-30percent-medium", | ||
| path=Path("action/action_classification/ucf_kinetics_30percent_medium"), | ||
| group="medium", | ||
| num_repeat=5, | ||
| extra_overrides={ | ||
| "train": { | ||
| "max_epochs": "10", | ||
| "deterministic": "True", | ||
| }, | ||
| }, | ||
| ), | ||
| Benchmark.Dataset( | ||
| name="ucf-large", | ||
| path=Path("action/action_classification/ucf_kinetics_large"), | ||
| group="large", | ||
| num_repeat=5, | ||
| extra_overrides={ | ||
| "train": { | ||
| "max_epochs": "3", | ||
| "deterministic": "True", | ||
| }, | ||
| }, | ||
| ), | ||
| ] | ||
|
|
||
| BENCHMARK_CRITERIA = [ # noqa: RUF012 | ||
| Benchmark.Criterion(name="train/epoch", summary="max", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="train/e2e_time", summary="max", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="test/accuracy", summary="max", compare=">", margin=0.1), | ||
| Benchmark.Criterion(name="export/accuracy", summary="max", compare=">", margin=0.1), | ||
| Benchmark.Criterion(name="optimize/accuracy", summary="max", compare=">", margin=0.1), | ||
| Benchmark.Criterion(name="train/iter_time", summary="mean", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="test/iter_time", summary="mean", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="export/iter_time", summary="mean", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="optimize/iter_time", summary="mean", compare="<", margin=0.1), | ||
| ] | ||
|
|
||
| @pytest.mark.parametrize( | ||
| "fxt_model", | ||
| MODEL_TEST_CASES, | ||
| ids=lambda model: model.name, | ||
| indirect=True, | ||
| ) | ||
| @pytest.mark.parametrize( | ||
| "fxt_dataset", | ||
| DATASET_TEST_CASES, | ||
| ids=lambda dataset: dataset.name, | ||
| indirect=True, | ||
| ) | ||
| def test_perf( | ||
| self, | ||
| fxt_model: Benchmark.Model, | ||
| fxt_dataset: Benchmark.Dataset, | ||
| fxt_benchmark: Benchmark, | ||
| ): | ||
| self._test_perf( | ||
| model=fxt_model, | ||
| dataset=fxt_dataset, | ||
| benchmark=fxt_benchmark, | ||
| criteria=self.BENCHMARK_CRITERIA, | ||
| ) | ||
|
|
||
|
|
||
| class TestPerfActionDetection(PerfTestBase): | ||
| """Benchmark action detection.""" | ||
|
|
||
| MODEL_TEST_CASES = [ # noqa: RUF012 | ||
| Benchmark.Model(task="action/action_detection", name="x3d_fastrcnn", category="accuracy"), | ||
| ] | ||
|
|
||
| DATASET_TEST_CASES = [ # noqa: RUF012 | ||
| Benchmark.Dataset( | ||
| name="ucf-5percent-small", | ||
| path=Path("action/action_detection/UCF101_ava_5percent"), | ||
| group="small", | ||
| num_repeat=5, | ||
| extra_overrides={ | ||
| "train": { | ||
| "max_epochs": "3", | ||
| "deterministic": "True", | ||
| }, | ||
| }, | ||
| ), | ||
| Benchmark.Dataset( | ||
| name="ucf-30percent-medium", | ||
| path=Path("action/action_detection/UCF101_ava_30percent"), | ||
| group="medium", | ||
| num_repeat=5, | ||
| extra_overrides={ | ||
| "train": { | ||
| "max_epochs": "3", | ||
| "deterministic": "True", | ||
| }, | ||
| }, | ||
| ), | ||
| Benchmark.Dataset( | ||
| name="ucf-large", | ||
| path=Path("action/action_detection/UCF101_ava"), | ||
| group="large", | ||
| num_repeat=5, | ||
| extra_overrides={ | ||
| "train": { | ||
| "max_epochs": "1", | ||
| "deterministic": "True", | ||
| }, | ||
| }, | ||
| ), | ||
| ] | ||
|
|
||
| BENCHMARK_CRITERIA = [ # noqa: RUF012 | ||
| Benchmark.Criterion(name="train/epoch", summary="max", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="train/e2e_time", summary="max", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="test/map_50", summary="max", compare=">", margin=0.1), | ||
| Benchmark.Criterion(name="export/map_50", summary="max", compare=">", margin=0.1), | ||
| Benchmark.Criterion(name="optimize/map_50", summary="max", compare=">", margin=0.1), | ||
| Benchmark.Criterion(name="train/iter_time", summary="mean", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="test/iter_time", summary="mean", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="export/iter_time", summary="mean", compare="<", margin=0.1), | ||
| Benchmark.Criterion(name="optimize/iter_time", summary="mean", compare="<", margin=0.1), | ||
| ] | ||
|
|
||
| @pytest.mark.parametrize( | ||
| "fxt_model", | ||
| MODEL_TEST_CASES, | ||
| ids=lambda model: model.name, | ||
| indirect=True, | ||
| ) | ||
| @pytest.mark.parametrize( | ||
| "fxt_dataset", | ||
| DATASET_TEST_CASES, | ||
| ids=lambda dataset: dataset.name, | ||
| indirect=True, | ||
| ) | ||
| def test_perf( | ||
| self, | ||
| fxt_model: Benchmark.Model, | ||
| fxt_dataset: Benchmark.Dataset, | ||
| fxt_benchmark: Benchmark, | ||
| ): | ||
| self._test_perf( | ||
| model=fxt_model, | ||
| dataset=fxt_dataset, | ||
| benchmark=fxt_benchmark, | ||
| criteria=self.BENCHMARK_CRITERIA, | ||
| ) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.