Merge pull request #277 from visual-layer/dnth/tests

Add tests folder structure and GitHub actions

Author: amiralush

.github/workflows/test-pypi-wheels.yml

@@ -0,0 +1,51 @@
+name: Daily Run Tests on PyPI Wheels
+
+on: 
+  workflow_dispatch:
+  schedule:
+    - cron: '0 16 * * *' # Trigger at 4PM every day
+
+jobs:
+    run_unit_tests:
+      runs-on: ${{ matrix.os }}
+      env:
+        SENTRY_OPT_OUT: True
+      strategy:
+        fail-fast: false
+        matrix:
+            os: [ubuntu-22.04] # windows-2022 macos-13
+            python: ['3.7', '3.8', '3.9','3.10']
+      steps:
+        - name: Checkout code
+          uses: actions/checkout@v3
+          with:
+           fetch-depth: 0
+  
+        - name: Set up Python
+          uses: actions/setup-python@v3
+          with:
+            python-version: ${{ matrix.python }}
+
+        - name: Install dependencies (Linux)
+          if: runner.os == 'Linux'
+          run: |
+            python -m pip install --upgrade pip
+            pip install fastdup pytest pytest-md pytest-emoji datasets
+          
+        - name: Install dependencies (macOS)
+          if: runner.os == 'macOS'
+          run: |
+            python -m pip install --upgrade pip
+            pip install pytest pytest-cov datasets fastdup
+            pip install opencv-python-headless
+            pip install fastdup --no-deps -U
+          
+        - name: Run pytest
+          uses: pavelzw/pytest-action@v2
+          with:
+            verbose: true
+            emoji: true
+            job-summary: true
+            custom-arguments: '-q'
+            click-to-expand: true
+            report-title: 'Test Report'

tests/README.md

@@ -0,0 +1,48 @@
+# Running Tests Locally
+
+To run the tests first install `pytest`:
+
+```
+pip install pytest
+```
+
+Then `cd` into the fastdup site-packages root directory. 
+
+For example
+
+```
+cd /home/dnth/anaconda3/envs/tests/lib/python3.10/site-packages/fastdup
+```
+
+and run
+
+```
+pytest
+```
+
+If all test passes, the output should look like the following:
+
+```shell
+========================================================================================== test session starts ===========================================================================================
+platform linux -- Python 3.10.13, pytest-7.4.2, pluggy-1.3.0
+rootdir: /media/dnth/Active-Projects/fastdup
+plugins: anyio-3.7.1, cov-4.1.0, mock-3.11.1
+collected 5 items                                                                                                                                                                                        
+
+tests/examples/test_datasets_example.py .                                                                                                                                                          [ 20%]
+tests/unit/test_datasets.py ....                                                                                                                                                                   [100%]
+
+============================================================================================ warnings summary ============================================================================================
+tests/examples/test_datasets_example.py::test_datasets_example
+  /home/dnth/anaconda3/envs/tests/lib/python3.10/site-packages/fastdup/fastdup_controller.py:385: UserWarning: No connected components found, try using a lower threshold
+    warnings.warn(f'No connected components found, try using a lower threshold')
+
+-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
+====================================================================================== 5 passed, 1 warning in 9.12s ======================================================================================
+```
+
+# Adding More Tests
+Include new tests in the approriate folders.
+
++ `unit/` - Put all unit tests here. 
++ `examples/` - Put all examples tests here.

tests/__init__.py

@@ -0,0 +1,7 @@
+import fastdup
+from fastdup.datasets import FastdupHFDataset
+
+def test_datasets_example():
+    dataset = FastdupHFDataset("zh-plus/tiny-imagenet", split="train[:5%]") # Download 5% of the train set
+    fd = fastdup.create(input_dir=dataset.img_dir)
+    fd.run(annotations=dataset.annotations, num_images=100)

tests/examples/test_datasets_example.py

@@ -0,0 +1,45 @@
+import pytest
+from fastdup.datasets import FastdupHFDataset
+from datasets.config import HF_DATASETS_CACHE
+import os
+import pandas as pd
+
+
+@pytest.fixture(scope="module")
+def dataset():
+    return FastdupHFDataset("zh-plus/tiny-imagenet", split="train[:5%]")
+
+
+def test_instantiating_dataset_with_valid_dataset_name_and_split(dataset):
+    assert isinstance(dataset, FastdupHFDataset)
+    assert dataset.img_key == "image"
+    assert dataset.label_key == "label"
+    assert dataset.jpg_save_dir == "jpg_images"
+    assert dataset.cache_dir == HF_DATASETS_CACHE
+    assert dataset.info.dataset_name == "tiny-imagenet"
+    assert len(dataset) == 5000
+
+
+def test_img_dir_exists(dataset):
+    assert os.path.isdir(dataset.img_dir)
+    assert dataset.img_dir == os.path.join(
+        HF_DATASETS_CACHE, "tiny-imagenet", "jpg_images"
+    )
+
+
+# Accessing img_dir property after successful instantiation
+def test_accessing_img_dir_property(dataset):
+    assert isinstance(dataset.img_dir, str)
+    assert os.path.exists(dataset.img_dir)
+
+
+# Accessing annotations property after successful instantiation
+def test_accessing_annotations_property_after_successful_instantiation(dataset):
+    annotations = dataset.annotations
+    assert isinstance(annotations, pd.DataFrame)
+    assert "filename" in annotations.columns
+    assert "label" in annotations.columns
+
+    sample_row = dataset.annotations.iloc[0]
+    assert os.path.exists(sample_row["filename"])
+    assert isinstance(sample_row["label"], str)