Upload 4 files

ViT-B-32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40d365715913c9da98579312b702a82c18be219cc2a73407c4526f58eba950af
+size 353976522

app.py

@@ -0,0 +1,72 @@
+# app.py
+import gradio as gr
+import requests
+from PIL import Image
+import io
+
+# --- Configuration ---
+BACKEND_URL = "http://127.0.0.1:8000/predict"
+
+# --- Interface Logic ---
+def predict_gender(image):
+    """
+    Sends an image to the FastAPI backend and returns the prediction.
+    'image' is a NumPy array from the Gradio Image component.
+    """
+    if image is None:
+        raise gr.Error("Please upload an image first.")
+
+    try:
+        # Convert numpy array to bytes
+        pil_image = Image.fromarray(image.astype('uint8'), 'RGB')
+        img_byte_arr = io.BytesIO()
+        pil_image.save(img_byte_arr, format='PNG')
+        img_byte_arr.seek(0) # Move cursor to the beginning of the buffer
+
+        # Prepare the file for the POST request
+        files = {'file': ('image.png', img_byte_arr, 'image/png')}
+
+        # Send request to the backend
+        response = requests.post(BACKEND_URL, files=files, timeout=30)
+        
+        # Process the response
+        if response.status_code == 200:
+            return response.json()
+        else:
+            # Display error from the backend as a Gradio error
+            error_detail = response.json().get('detail', 'An unknown error occurred.')
+            raise gr.Error(f"API Error: {error_detail}")
+
+    except requests.exceptions.RequestException as e:
+        raise gr.Error(f"Could not connect to the backend. Please ensure the backend is running. Details: {e}")
+    except Exception as e:
+        raise gr.Error(f"An unexpected error occurred: {e}")
+
+
+# --- Gradio Interface Definition ---
+iface = gr.Interface(
+    fn=predict_gender,
+    inputs=gr.Image(label="Upload a Photo", type="numpy"),
+    outputs=gr.Label(label="Gender Prediction", num_top_classes=2),
+    title="📸 Gender Prediction with CLIP",
+    description=(
+        "Upload a clear, front-facing photo of a single person to predict their gender. "
+        "The app uses a backend API powered by OpenAI's CLIP model."
+    ),
+    examples=[
+        ["examples/male_example.jpg"],
+        ["examples/female_example.jpg"],
+    ],
+    allow_flagging="never",
+    css=".gradio-container {max-width: 780px !important; margin: auto;}"
+)
+
+# --- Launch the App ---
+if __name__ == "__main__":
+    # Create an 'examples' directory for Gradio examples if it doesn't exist
+    import os
+    if not os.path.exists("examples"):
+        os.makedirs("examples")
+        print("Created 'examples' directory. Please add 'male_example.jpg' and 'female_example.jpg' for the demo.")
+    
+    iface.launch()

main.py

@@ -0,0 +1,121 @@
+# main.py
+import uvicorn
+import numpy as np
+import clip
+import torch
+from fastapi import FastAPI, File, UploadFile, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from retinaface import RetinaFace
+from PIL import Image
+import io
+import os
+
+# --- Constants & Configuration ---
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+MODELS_DIR = "models"
+GENDER_PROMPTS = ["a photo of a man", "a photo of a woman"]
+
+# --- Error Messages ---
+ERROR_MESSAGES = {
+    "NO_FACE": "No face detected. Please upload a clear, front-facing picture of a single person.",
+    "MULTIPLE_FACES": "Multiple faces detected. Please upload an image with only one face.",
+    "ANALYSIS_ERROR": "An unexpected error occurred during analysis. Please try again.",
+    "FILE_READ_ERROR": "Could not read the uploaded file. Please ensure it's a valid image."
+}
+
+# --- Model Loading ---
+# Create models directory if it doesn't exist
+os.makedirs(MODELS_DIR, exist_ok=True)
+
+try:
+    print(f"Loading CLIP model on device: {DEVICE}...")
+    # Load the model, downloading to the specified directory if necessary
+    model, preprocess = clip.load("ViT-B/32", device=DEVICE, download_root=MODELS_DIR)
+    print("✓ CLIP model loaded successfully.")
+except Exception as e:
+    print(f"✗ Failed to load CLIP model: {e}")
+    exit()
+
+# --- FastAPI App Initialization ---
+app = FastAPI(
+    title="Gender Detection API",
+    description="A simple API using CLIP to predict gender from an image."
+)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # Allows all origins for simplicity
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# --- Core Logic ---
+def predict_gender_with_clip(image: Image.Image) -> dict:
+    """
+    Predicts gender from a PIL Image using the loaded CLIP model.
+
+    Args:
+        image (Image.Image): The input image.
+
+    Returns:
+        dict: A dictionary with gender labels and their confidence scores.
+    """
+    image_input = preprocess(image).unsqueeze(0).to(DEVICE)
+    text_inputs = clip.tokenize(GENDER_PROMPTS).to(DEVICE)
+
+    with torch.no_grad():
+        logits_per_image, _ = model(image_input, text_inputs)
+        # Softmax to get probabilities
+        probabilities = logits_per_image.softmax(dim=-1).cpu().numpy()[0]
+
+    # Map probabilities to labels
+    return {GENDER_PROMPTS[i].split("of a ")[-1]: float(prob) for i, prob in enumerate(probabilities)}
+
+
+# --- API Endpoints ---
+@app.get("/health")
+async def health_check():
+    """Health check endpoint to verify if the API is running."""
+    return {"status": "healthy"}
+
+@app.post("/predict")
+async def predict(file: UploadFile = File(...)):
+    """
+    Main prediction endpoint. It validates the image and returns gender probabilities.
+    """
+    try:
+        # 1. Read and validate the uploaded image
+        contents = await file.read()
+        image_pil = Image.open(io.BytesIO(contents)).convert("RGB")
+        # Convert to numpy array for face detection (expects BGR)
+        image_np = np.array(image_pil)
+        image_np = image_np[:, :, ::-1].copy()  # RGB -> BGR
+    except Exception:
+        raise HTTPException(status_code=400, detail=ERROR_MESSAGES["FILE_READ_ERROR"])
+
+    try:
+        # 2. Detect faces using RetinaFace
+        faces = RetinaFace.detect_faces(image_np)
+        num_faces = len(faces)
+
+        if num_faces == 0:
+            raise HTTPException(status_code=422, detail=ERROR_MESSAGES["NO_FACE"])
+        if num_faces > 1:
+            raise HTTPException(status_code=422, detail=ERROR_MESSAGES["MULTIPLE_FACES"])
+        
+        # 3. Predict gender using CLIP
+        gender_probabilities = predict_gender_with_clip(image_pil)
+        
+        return gender_probabilities
+
+    except HTTPException as e:
+        # Re-raise known HTTP exceptions
+        raise e
+    except Exception as e:
+        print(f"An unexpected error occurred: {e}")
+        raise HTTPException(status_code=500, detail=ERROR_MESSAGES["ANALYSIS_ERROR"])
+
+# --- Main Execution ---
+if __name__ == "__main__":
+    uvicorn.run(app, host="127.0.0.1", port=8000)

requirements.txt

@@ -0,0 +1,93 @@
+absl-py==2.3.0
+annotated-types==0.7.0
+anyio==4.9.0
+astunparse==1.6.3
+attrs==25.3.0
+beautifulsoup4==4.13.4
+cachetools==5.5.2
+certifi==2025.6.15
+cffi==1.17.1
+charset-normalizer==3.4.2
+click==8.2.1
+clip @ git+https://github.com/openai/CLIP.git@dcba3cb2e2827b402d2701e7e1c7d9fed8a20ef1
+colorama==0.4.6
+contourpy==1.3.2
+cycler==0.12.1
+exceptiongroup==1.3.0
+fastapi==0.115.13
+filelock==3.18.0
+flatbuffers==25.2.10
+fonttools==4.58.4
+fsspec==2025.5.1
+ftfy==6.3.1
+gast==0.6.0
+gdown==5.2.0
+google-auth==2.40.3
+google-auth-oauthlib==1.0.0
+google-pasta==0.2.0
+grpcio==1.73.1
+h11==0.16.0
+h5py==3.14.0
+idna==3.10
+jax==0.4.34
+jaxlib==0.4.34
+jinja2==3.1.6
+keras==2.14.0
+kiwisolver==1.4.8
+libclang==18.1.1
+markdown==3.8.2
+markupsafe==3.0.2
+matplotlib==3.10.3
+mediapipe==0.10.14
+ml-dtypes==0.2.0
+mpmath==1.3.0
+networkx==3.4.2
+numpy==1.24.4
+oauthlib==3.3.1
+opencv-contrib-python==4.11.0.86
+opencv-python==4.11.0.86
+opencv-python-headless==4.11.0.86
+opt-einsum==3.4.0
+packaging==25.0
+pillow==11.2.1
+protobuf==4.25.8
+pyasn1==0.6.1
+pyasn1-modules==0.4.2
+pycparser==2.22
+pydantic==2.11.7
+pydantic-core==2.33.2
+pyparsing==3.2.3
+pysocks==1.7.1
+python-dateutil==2.9.0.post0
+regex==2024.11.6
+requests==2.32.4
+requests-oauthlib==2.0.0
+retina-face==0.0.17
+rsa==4.9.1
+scipy==1.15.3
+setuptools==80.9.0
+six==1.17.0
+sniffio==1.3.1
+sounddevice==0.5.2
+soupsieve==2.7
+starlette==0.46.2
+sympy==1.14.0
+tensorboard==2.14.1
+tensorboard-data-server==0.7.2
+tensorflow==2.14.0
+tensorflow-estimator==2.14.0
+tensorflow-intel==2.14.0
+tensorflow-io-gcs-filesystem==0.31.0
+termcolor==3.1.0
+torch==2.7.1
+torchaudio==2.7.1
+torchvision==0.22.1
+tqdm==4.67.1
+typing-extensions==4.14.0
+typing-inspection==0.4.1
+urllib3==2.5.0
+uvicorn==0.34.3
+wcwidth==0.2.13
+werkzeug==3.1.3
+wheel==0.45.1
+wrapt==1.14.1