Add application file

Dockerfile

@@ -0,0 +1,18 @@
+# Use an official lightweight Python image
+FROM python:3.10
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy required files
+COPY requirements.txt .
+COPY app.py .
+
+# Install dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Expose the API port
+EXPOSE 7860
+
+# Run FastAPI application
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -0,0 +1,96 @@
+import torch
+import torch.nn as nn
+from torchvision import transforms
+from PIL import Image
+import cv2
+import timm
+from fastapi import FastAPI, File, UploadFile
+import shutil
+import os
+from huggingface_hub import hf_hub_download
+
+# FastAPI app instance
+app = FastAPI()
+
+# Device configuration
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+print("Initializing model...")
+
+# Define image transformations
+transform = transforms.Compose([
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
+])
+
+# Hugging Face model repository details
+HF_MODEL_REPO = "Maddy21/vit-deepfake-model"  # Replace with your repo
+HF_MODEL_FILE = "best_vit_model.pth"
+
+# Download the model from Hugging Face
+model_path = hf_hub_download(repo_id=HF_MODEL_REPO, filename=HF_MODEL_FILE)
+
+# Load the trained ViT model
+model = timm.create_model('vit_large_patch16_224', pretrained=False, num_classes=2)
+model.load_state_dict(torch.load(model_path, map_location=device))
+model.to(device)
+model.eval()
+print("Model loaded successfully.")
+
+# Function to process video and classify frames
+def predict_video(video_path):
+    print("Processing video for deepfake detection...")
+    cap = cv2.VideoCapture(video_path)
+    frame_count = 0
+    real_count = 0
+    manipulated_count = 0
+
+    while cap.isOpened():
+        ret, frame = cap.read()
+        if not ret:
+            break
+        
+        frame_count += 1
+        image = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB))
+        image = transform(image).unsqueeze(0).to(device)
+
+        with torch.no_grad():
+            outputs = model(image)
+            _, predicted = torch.max(outputs, 1)
+        
+        if predicted.item() == 0:
+            real_count += 1
+        else:
+            manipulated_count += 1
+
+    cap.release()
+
+    result = "Real" if real_count > manipulated_count else "Manipulated"
+    return {
+        "total_frames": frame_count,
+        "real_frames": real_count,
+        "manipulated_frames": manipulated_count,
+        "result": result
+    }
+
+# API Endpoint to check if API is running
+@app.get("/")
+def read_root():
+    return {"message": "Deepfake Detection API is running!"}
+
+# API Endpoint to upload a video and get predictions
+@app.post("/predict/")
+async def predict(file: UploadFile = File(...)):
+    file_path = f"temp_{file.filename}"
+    
+    # Save uploaded video
+    with open(file_path, "wb") as buffer:
+        shutil.copyfileobj(file.file, buffer)
+    
+    # Run prediction
+    result = predict_video(file_path)
+
+    # Delete temp file after processing
+    os.remove(file_path)
+    return result
+