Initial commit

.gitignore

@@ -0,0 +1,165 @@
+# Environment file
+config_api.env
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+.idea/

app.py

@@ -0,0 +1,116 @@
+import gradio as gr
+
+import os
+import requests
+import json
+import utils
+
+from dotenv import load_dotenv, find_dotenv
+
+# List of ML models
+list_models = ["facebook/detr-resnet-50", "facebook/detr-resnet-101", "hustvl/yolos-tiny", "hustvl/yolos-small"]
+list_models_simple = [os.path.basename(model) for model in list_models]
+
+# ECS APIs
+AWS_DETR_URL = None
+AWS_YOLOS_URL = None
+
+
+# Initialize API URLs from env file or global settings
+def initialize_api_endpoints():
+
+    env_path = find_dotenv('config_api.env')
+    if env_path:
+        load_dotenv(dotenv_path=env_path)
+        print("config_api.env file loaded successfully.")
+    else:
+        print("config_api.env file not found.")
+
+    # Use of AWS ECS endpoint or local container by default
+    global AWS_DETR_URL,  AWS_YOLOS_URL
+    AWS_DETR_URL = os.getenv("AWS_DETR_URL", default="http://0.0.0.0:8000")
+    AWS_YOLOS_URL = os.getenv("AWS_YOLOS_URL", default="http://0.0.0.0:8001")
+
+
+# Retrieve correct endpoint based on model_type
+def retrieve_api_endpoint(model_type):
+    if "detr" in model_type:
+        API_URL = AWS_DETR_URL
+    else:
+        API_URL = AWS_YOLOS_URL
+
+    return API_URL
+
+
+#@spaces.GPU
+def detect(image_path, model_id, threshold):
+    print("\n Object detection...")
+    print("\t ML model:", list_models[model_id])
+
+    with open(image_path, 'rb') as image_file:
+       image_bytes = image_file.read()
+
+    API_URL = retrieve_api_endpoint(list_models_simple[model_id])
+
+    # API Call for object prediction with model type as query parameter
+    API_Endpoint = API_URL + "/api/v1/detect" + "?model=" + list_models_simple[model_id]
+    print("\t API_Endpoint: ", API_Endpoint)
+
+    response = requests.post(API_Endpoint, files={"image": image_bytes})
+    if response.status_code == 200:
+        # Process the response
+        response_string = response.json()
+        response_dict = json.loads(response_string)
+        print('\t API response', response_string)
+    else:
+        response_dict = {"Error": response.status_code}
+        gr.Error(f"\t API Error: {response.status_code}")
+
+    # Generate gradio output components: image and json
+    output_json, output_pil_img = utils.generate_gradio_outputs(image_path, response_dict, threshold)
+
+    return output_json, output_pil_img
+
+
+def demo():
+    initialize_api_endpoints()
+    with gr.Blocks(theme="base") as demo:
+        gr.Markdown("# Object detection task - use of ECS endpoints")
+        gr.Markdown(
+            """
+            This web application uses transformer models to detect objects on images.
+            Machine learning models were trained on the COCO dataset.
+            You can load an image and see the predictions for the objects detected.
+            
+            Note: This web application uses AWS ECS endpoints as a back-end APIs to run these ML models.
+            """
+        )
+
+        with gr.Row():
+            with gr.Column():
+                model_id = gr.Radio(list_models, \
+                               label="Detection models", value=list_models[0], type="index", info="Choose your detection model")
+            with gr.Column():
+                threshold = gr.Slider(0, 1.0, value=0.9, label='Detection threshold', info="Choose your detection threshold")
+
+        with gr.Row():
+            input_image = gr.Image(label="Input image", type="filepath")
+            output_image = gr.Image(label="Output image", type="pil")
+            output_json = gr.JSON(label="JSON output", min_height=240, max_height=300)
+
+        with gr.Row():
+            submit_btn = gr.Button("Submit")
+            clear_button = gr.ClearButton()
+
+        gr.Examples(['samples/savanna.jpg', 'samples/boats.jpg'], inputs=input_image)
+
+        submit_btn.click(fn=detect, inputs=[input_image, model_id, threshold], outputs=[output_json, output_image])
+        clear_button.click(lambda: [None, None, None], \
+                        inputs=None, \
+                        outputs=[input_image, output_image, output_json], \
+                        queue=False)
+
+    demo.queue().launch(debug=True)
+
+if __name__ == "__main__":
+    demo()

utils.py

@@ -0,0 +1,86 @@
+from PIL import Image
+import matplotlib.pyplot as plt
+import io
+
+
+# COCO classes
+CLASSES = [
+    'N/A', 'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus',
+    'train', 'truck', 'boat', 'traffic light', 'fire hydrant', 'N/A',
+    'stop sign', 'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse',
+    'sheep', 'cow', 'elephant', 'bear', 'zebra', 'giraffe', 'N/A', 'backpack',
+    'umbrella', 'N/A', 'N/A', 'handbag', 'tie', 'suitcase', 'frisbee', 'skis',
+    'snowboard', 'sports ball', 'kite', 'baseball bat', 'baseball glove',
+    'skateboard', 'surfboard', 'tennis racket', 'bottle', 'N/A', 'wine glass',
+    'cup', 'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple', 'sandwich',
+    'orange', 'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake',
+    'chair', 'couch', 'potted plant', 'bed', 'N/A', 'dining table', 'N/A',
+    'N/A', 'toilet', 'N/A', 'tv', 'laptop', 'mouse', 'remote', 'keyboard',
+    'cell phone', 'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'N/A',
+    'book', 'clock', 'vase', 'scissors', 'teddy bear', 'hair drier',
+    'toothbrush'
+]
+COLORS = [
+    [0.000, 0.447, 0.741],
+    [0.850, 0.325, 0.098],
+    [0.929, 0.694, 0.125],
+    [0.494, 0.184, 0.556],
+    [0.466, 0.674, 0.188],
+    [0.301, 0.745, 0.933],
+]
+
+
+# Update JSON dictionary with rounded values and classes
+def generate_output_json(json_dict):
+    json_dict['scores'] = [round(score, 3) for score in json_dict['scores']]
+    json_dict['boxes'] = [[round(coord, 3) for coord in box] for box in json_dict['boxes']]
+    json_dict['labels'] = [CLASSES[label] for label in json_dict['labels']]
+    return json_dict
+
+
+# Generate matplotlib figure from prediction scores and boxes
+def generate_output_figure(image_path, results, threshold):
+    pil_img = Image.open(image_path)
+
+    plt.figure(figsize=(16, 10))
+    plt.imshow(pil_img)
+    ax = plt.gca()
+    colors = COLORS * 100
+
+    print("\t Detailed information...")
+    for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
+        #box = [round(i, 2) for i in box]
+        print(
+            f"\t\t Detected {label} with confidence "
+            f"{score} at location {box}"
+        )
+
+        if score > threshold:
+            c = COLORS[hash(label) % len(COLORS)]
+            ax.add_patch(
+                plt.Rectangle((box[0], box[1]), box[2] - box[0], box[3] - box[1], fill=False, color=c, linewidth=3)
+            )
+            text = f"{label}: {score:0.2f}"
+            ax.text(box[0], box[1], text, fontsize=15, bbox=dict(facecolor="yellow", alpha=0.5))
+    plt.axis("off")
+
+    return plt.gcf()
+
+
+#  Generate PIL image from matplotlib figure
+def generate_output_image(output_figure):
+    # Convert matplotlib figure to PIL image
+    #output_figure = plt.gcf()
+    buf = io.BytesIO()
+    output_figure.savefig(buf, bbox_inches="tight")
+    buf.seek(0)
+    output_pil_img = Image.open(buf)
+
+    return output_pil_img
+
+
+def generate_gradio_outputs(image_path, response_dict, threshold):
+    output_json = generate_output_json(response_dict)
+    output_figure = generate_output_figure(image_path, output_json, threshold)
+    output_pil_img = generate_output_image(output_figure)
+    return output_json, output_pil_img