initial commit

app/main.py

@@ -0,0 +1,51 @@
+# app/main.py
+
+import os
+from fastapi import FastAPI, UploadFile, File, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import JSONResponse
+from app.utils.image_utils import load_model, process_image
+import torch
+
+# Initialize FastAPI app
+app = FastAPI()
+
+# CORS settings for frontend communication
+# Add CORS middleware to allow requests from your frontend
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["http://localhost:5173","https://eye-disease-detection-frontend.vercel.app/"],  # Set your frontend URL here
+    allow_credentials=True,
+    allow_methods=["*"],  # Allow all HTTP methods (GET, POST, etc.)
+    allow_headers=["*"],  # Allow all headers
+)
+
+# Model and label setup
+MODEL_PATH = r"D:\College Works\ML_project\Web\api\app\model\eye_disease_model.pth"
+categories = ["cataract", "diabetic_retinopathy", "glaucoma", "normal"]
+
+try:
+    model = load_model(MODEL_PATH)
+except FileNotFoundError as e:
+    print(e)
+    model = None
+
+# Prediction endpoint
+@app.post("/predict")
+async def predict(file: UploadFile = File(...)):
+    if model is None:
+        raise HTTPException(status_code=500, detail="Model not loaded")
+
+    contents = await file.read()
+    
+    try:
+        image = process_image(contents)
+    except Exception as e:
+        raise HTTPException(status_code=400, detail=f"Invalid image: {e}")
+    
+    with torch.no_grad():
+        outputs = model(image)
+        _, predicted = torch.max(outputs, 1)
+        prediction = categories[predicted.item()]
+
+    return JSONResponse(content={"prediction": prediction})

app/model/eye_disease_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27a2e05d406289e1ac51649c2605d30d32cf917a3c1a9044e03aa8bcbb35f42a
+size 45845690

app/utils/image_utils.py

@@ -0,0 +1,53 @@
+import torch
+from torchvision import transforms, models
+from PIL import Image
+import io
+import os
+import torch.nn as nn
+
+# Model path
+MODEL_PATH = r"D:\College Works\ML_project\Web\api\app\model\eye_disease_model.pth"
+
+# Load model
+def load_model(model_path=MODEL_PATH):
+    if not os.path.exists(model_path):  # Check if model path exists
+        raise FileNotFoundError(f"Model file not found at: {model_path}")
+    
+    # Define the model (with custom layers as per your training)
+    model = models.resnet18(pretrained=False)
+    
+    # Freeze the layers and modify the final layers
+    for param in model.parameters():
+        param.requires_grad = False
+        
+    num_ftrs = model.fc.in_features
+    model.fc = nn.Sequential(
+        nn.Linear(num_ftrs, 512),
+        nn.ReLU(),
+        nn.Dropout(0.5),
+        nn.Linear(512, 4)  # Assuming 4 output classes for your eye disease model
+    )
+
+    # Load the trained weights into the model
+    model.load_state_dict(torch.load(model_path, map_location=torch.device("cpu")))
+    
+    # Set the model to evaluation mode
+    model.eval()
+    
+    return model
+
+# Define image transformation (should match test-time transforms)
+def get_transform():
+    return transforms.Compose([
+        transforms.Resize((224, 224)),
+        transforms.ToTensor(),
+        transforms.Normalize(mean=[0.485, 0.456, 0.406],
+                             std=[0.229, 0.224, 0.225])
+    ])
+
+# Process the incoming image
+def process_image(contents):
+    image = Image.open(io.BytesIO(contents)).convert("RGB")
+    transform = get_transform()
+    image = transform(image).unsqueeze(0)  # Add batch dimension
+    return image

model/eye_disease_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27a2e05d406289e1ac51649c2605d30d32cf917a3c1a9044e03aa8bcbb35f42a
+size 45845690

requirements.txt

@@ -0,0 +1,27 @@
+annotated-types==0.7.0
+anyio==4.9.0
+click==8.1.8
+colorama==0.4.6
+exceptiongroup==1.2.2
+fastapi==0.115.12
+filelock==3.18.0
+fsspec==2025.3.2
+h11==0.16.0
+idna==3.10
+Jinja2==3.1.6
+MarkupSafe==3.0.2
+mpmath==1.3.0
+networkx==3.2.1
+numpy==2.0.2
+pillow==11.2.1
+pydantic==2.11.4
+pydantic_core==2.33.2
+python-multipart==0.0.20
+sniffio==1.3.1
+starlette==0.46.2
+sympy==1.14.0
+torch==2.7.0
+torchvision==0.22.0
+typing-inspection==0.4.0
+typing_extensions==4.13.2
+uvicorn==0.34.2

utils/image_utils.py

@@ -0,0 +1,53 @@
+import torch
+from torchvision import transforms, models
+from PIL import Image
+import io
+import os
+import torch.nn as nn
+
+# Model path
+MODEL_PATH = r"D:\College Works\ML_project\Web\api\app\model\eye_disease_model.pth"
+
+# Load model
+def load_model(model_path=MODEL_PATH):
+    if not os.path.exists(model_path):  # Check if model path exists
+        raise FileNotFoundError(f"Model file not found at: {model_path}")
+    
+    # Define the model (with custom layers as per your training)
+    model = models.resnet18(pretrained=False)
+    
+    # Freeze the layers and modify the final layers
+    for param in model.parameters():
+        param.requires_grad = False
+        
+    num_ftrs = model.fc.in_features
+    model.fc = nn.Sequential(
+        nn.Linear(num_ftrs, 512),
+        nn.ReLU(),
+        nn.Dropout(0.5),
+        nn.Linear(512, 4)  # Assuming 4 output classes for your eye disease model
+    )
+
+    # Load the trained weights into the model
+    model.load_state_dict(torch.load(model_path, map_location=torch.device("cpu")))
+    
+    # Set the model to evaluation mode
+    model.eval()
+    
+    return model
+
+# Define image transformation (should match test-time transforms)
+def get_transform():
+    return transforms.Compose([
+        transforms.Resize((224, 224)),
+        transforms.ToTensor(),
+        transforms.Normalize(mean=[0.485, 0.456, 0.406],
+                             std=[0.229, 0.224, 0.225])
+    ])
+
+# Process the incoming image
+def process_image(contents):
+    image = Image.open(io.BytesIO(contents)).convert("RGB")
+    transform = get_transform()
+    image = transform(image).unsqueeze(0)  # Add batch dimension
+    return image