feat: adding integration test

.github/workflows/ci.yaml

@@ -32,4 +32,7 @@ jobs:
           poetry install
 
       - name: Run Pre commit hooks
-        run: make format-code
+        run: make format-code
+
+      - name: Run Tests
+        run: poetry run pytest tests

README.md

@@ -42,3 +42,10 @@ ByteTrack is licensed under the MIT License. See the [LICENSE](LICENSE) file and
   year={2022}
 }
 ```
+
+```bibtex
+@video{intel-iot-devkit,
+  author={Intel IoT},
+  title={Intel IoT Devkit Sample Videos(https://github.com/intel-iot-devkit/sample-videos)},
+  license={Creative Commons Attribution 4.0 International License(https://creativecommons.org/licenses/by/4.0/)}
+```

tests/expected_output/objects_detected_and_tracked_video2.txt

@@ -0,0 +1,19 @@
+117.0 0.974591305835963 0.5095593273639679 0.9995998740196228 0.742304390668869 142.0 49.0 0.8319151401519775
+118.0 0.9723771181236184 0.5005956317592295 0.9995706969396306 0.745516412561992 142.0 49.0 0.8575639128684998
+119.0 0.9692141222864277 0.4963345558666765 0.9995033249395253 0.7480595143898368 142.0 49.0 0.8860840201377869
+120.0 0.9597183272620015 0.49863389352533755 0.9950719974612193 0.75323887234266 142.0 49.0 0.9059332013130188
+121.0 0.9137051296973745 0.49948144088198765 0.9549359841485002 0.7572477348922191 142.0 49.0 0.9214490652084351
+122.0 0.8638724059525613 0.49995341031172713 0.9114848810392273 0.7600943052013691 142.0 49.0 0.9317426085472107
+123.0 0.8202443010481131 0.5008927321202477 0.8744436588068774 0.7617456617798124 142.0 49.0 0.9288402199745178
+124.0 0.7771122186769683 0.500385486071895 0.8379277714479001 0.7611541221850968 142.0 49.0 0.9290407299995422
+125.0 0.7247772874637586 0.5017298121516314 0.7923321383198036 0.76303004156558 142.0 49.0 0.9264274835586548
+126.0 0.6671303227010921 0.5027124081353996 0.7416950553044594 0.7661639362088702 142.0 49.0 0.9315325617790222
+127.0 0.616663571672274 0.5032080911861249 0.6977953166390259 0.7674052629688627 142.0 49.0 0.9363259673118591
+128.0 0.5704205150806596 0.5034447554058634 0.6576470426224459 0.7674596817474673 142.0 49.0 0.9300203919410706
+129.0 0.5200086795843045 0.5042046723093252 0.6129952764525678 0.7679312663000262 142.0 49.0 0.927420973777771
+130.0 0.46643053919994637 0.5047469314997519 0.5650969063243967 0.7687824676857495 142.0 49.0 0.9283919334411621
+131.0 0.4173761495829712 0.5047965664538961 0.5216674559275042 0.7698882380253113 142.0 49.0 0.9305659532546997
+132.0 0.36950273408788886 0.5046674217427534 0.47893139831600645 0.770454233806388 142.0 49.0 0.922694206237793
+133.0 0.3186225320703163 0.5049336319091298 0.43268114879297337 0.7708076409122564 142.0 49.0 0.9281331896781921
+134.0 0.26800115818737746 0.5048401786190908 0.38642683784144416 0.7709888863166645 142.0 49.0 0.9294517636299133
+135.0 0.22018770537261864 0.5043750340652045 0.3428432549812343 0.770896006411032 142.0 49.0 0.9294649362564087

tests/test_input/objects_detected_video2.txt

@@ -0,0 +1,21 @@
+40 0.21998224 0.5836777 0.3745204 0.7639922 0.16180652 49.0
+116 0.9795788 0.53099704 0.9996632 0.7194981 0.51681966 49.0
+117 0.95138013 0.5041999 0.999584 0.748006 0.83191514 49.0
+118 0.9269222 0.49708003 0.9995592 0.7467762 0.8575639 49.0
+119 0.88959634 0.49609375 0.9994829 0.7485177 0.886084 49.0
+120 0.8366054 0.50106037 0.99332654 0.75461787 0.9059332 49.0
+121 0.7823308 0.5008052 0.93827456 0.7578524 0.92144907 49.0
+122 0.73994255 0.50089943 0.89590263 0.76007366 0.9317426 49.0
+123 0.70718193 0.5019245 0.86416876 0.76115876 0.9288402 49.0
+124 0.6728785 0.5005684 0.8293528 0.75955844 0.92904073 49.0
+125 0.624425 0.50280184 0.78042 0.76286423 0.9264275 49.0
+126 0.5726967 0.5034708 0.72873324 0.7666482 0.93153256 49.0
+127 0.53285456 0.50363165 0.6897488 0.7668706 0.93632597 49.0
+128 0.4952455 0.5036848 0.6524906 0.7663839 0.9300204 49.0
+129 0.44878444 0.50468314 0.60615635 0.7671972 0.927421 49.0
+130 0.39896855 0.5050585 0.5573927 0.76832783 0.92839193 49.0
+131 0.35776815 0.5048203 0.51712275 0.76961535 0.93056595 49.0
+132 0.31567994 0.5045963 0.4752435 0.7699328 0.9226942 49.0
+133 0.26709083 0.5050746 0.4275921 0.7702349 0.9281332 49.0
+134 0.22100373 0.5047798 0.3819117 0.77039224 0.92945176 49.0
+135 0.17729278 0.5041295 0.3402054 0.77022386 0.92946494 49.0

tests/test_model_regression.py

@@ -0,0 +1,104 @@
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+# bytetracker
+from bytetracker import BYTETracker
+
+# static paths
+OUTPUT_FOLDER = Path("tests/output")
+TEST_INPUT_FOLDER = Path("tests/test_input")
+EXPECTED_OUTPUT_FOLDER = Path("tests/expected_output")
+OUTPUT_FOLDER = Path("tests/output")
+
+OUTPUT_FOLDER.mkdir(parents=True, exist_ok=True)
+
+
+def reading_detections_file(video_number):
+    """Read the detections object file from a specific format
+
+    Args:
+        video_number (string): string representing the video name in folder
+
+    Returns:
+        df_detection (ndarray): A list of array of the tuples.
+    """
+    df_detection = np.loadtxt(
+        TEST_INPUT_FOLDER / f"objects_detected_{video_number}.txt", delimiter=" "
+    )
+    print(df_detection)
+
+    return df_detection
+
+
+def reading_expected_results_from_txt(video_number):
+    """Read the detections and tracked objects frames from Yolo and Bytetrack models
+
+    Args:
+        video_number (str): Video title to import
+
+    Returns:
+        expected_results_df (ndarray): array of frames detection and tracking imported from expected results.
+    """
+    expected_results_df = np.loadtxt(
+        EXPECTED_OUTPUT_FOLDER / f"objects_detected_and_tracked_{video_number}.txt"
+    )
+
+    return expected_results_df
+
+
+@pytest.mark.parametrize(
+    "expected_results, test_input, video_number",
+    [
+        (
+            reading_expected_results_from_txt("video1"),
+            reading_detections_file("video1"),
+            "video1",
+        ),
+        (
+            reading_expected_results_from_txt("video2"),
+            reading_detections_file("video2"),
+            "video2",
+        ),
+    ],
+)
+def test_video_prediction_tracking(expected_results, test_input, video_number):
+    """Execute the integration test: performing the tracking of test_input,
+    whose output is to compare with expected_results
+
+    Args:
+        expected_result (DataFrame): the dataframe of the results expected to match the test result
+        test_input (array): dataframe that is the import of prediction frames in test_input folder
+        video_number (string): string component specific to a video
+
+    Returns:
+        Test assertion results
+    """
+    tracker = BYTETracker(track_thresh=0.15, track_buffer=3, match_thresh=0.85, frame_rate=12)
+    test_results = []
+    # first column of test input is the frame id
+    frame_idx = np.unique(test_input[:, 0])
+    test_results = []
+
+    for frame_id in frame_idx:
+        detections = test_input[test_input[:, 0] == frame_id][:, 1:]
+        tracked_objects = tracker.update(detections, frame_id)
+        if len(tracked_objects) > 0:
+            tracked_objects = np.insert(
+                tracked_objects, 0, np.full(len(tracked_objects), frame_id), axis=1
+            )
+            test_results.append(tracked_objects)
+
+    if test_results:  # Ensure test_results is not empty to avoid errors
+        combined_array = np.concatenate(test_results)
+    else:
+        combined_array = np.array([])  # Or handle the empty case as needed
+
+    output_file_path = OUTPUT_FOLDER / f"{video_number}_test_results.txt"
+    np.savetxt(output_file_path, combined_array, delimiter=" ", fmt="%s")
+
+    assert np.allclose(expected_results, combined_array)
+
+    # Remove the file if the test is successful
+    output_file_path.unlink()