Update helper.py

helper.py

@@ -0,0 +1,57 @@
+import torch
+from torchvision import transforms
+from PIL import Image
+import io
+
+MODEL_PATH = "model_checkpoint.pt"
+NUM_CLASSES = 4  
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+
+# Load Faster R-CNN model
+def load_model(model_path, num_classes):
+    from torchvision.models.detection import fasterrcnn_resnet50_fpn
+    model = fasterrcnn_resnet50_fpn(pretrained=False, num_classes=num_classes)
+    checkpoint = torch.load(model_path, map_location=DEVICE)
+    model.load_state_dict(checkpoint["model_state_dict"])
+    model.to(DEVICE)
+    model.eval()
+    return model
+
+transform = transforms.Compose([
+    transforms.Resize((640, 640)),
+    transforms.ToTensor(),
+])
+
+model = load_model(MODEL_PATH, NUM_CLASSES)
+
+def detect_objects(image_bytes):
+    image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
+    input_tensor = transform(image).unsqueeze(0).to(DEVICE)
+
+    with torch.no_grad():
+        predictions = model(input_tensor)
+
+    boxes = predictions[0]["boxes"].cpu().tolist()
+    labels = predictions[0]["labels"].cpu().tolist()
+    scores = predictions[0]["scores"].cpu().tolist()
+
+    confidence_threshold = 0.5
+    results = [
+        {"box": box, "label": label, "score": score}
+        for box, label, score in zip(boxes, labels, scores)
+        if score > confidence_threshold
+    ]
+
+    return {"predictions": results}
+
+def inference(payload):
+    try:
+        if "image" not in payload:
+            return {"error": "No image provided. Please send an image."}
+        
+        image_bytes = payload["image"].encode("latin1")  
+        
+        results = detect_objects(image_bytes)
+        return results
+    except Exception as e:
+        return {"error": str(e)}