Update handler.py

handler.py

@@ -1,12 +1,6 @@
-#from subprocess import run
-#run("pip install ultralytics", shell=True, check=True)
-#run("pip install opencv-python", shell=True, check=True)
 from huggingface_hub import hf_hub_download
 from typing import Dict, List, Any
 from ultralytics import YOLO
-#import cv2
-#import torch
-#import numpy as np
 
 class EndpointHandler():
     def __init__(self, path=""):
@@ -40,78 +34,4 @@ class EndpointHandler():
             return {
                 "statusCode": 500,
                 "body": json.dumps(f"Error: {str(e)}"),
-            }
-            
-    """
-    def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:    
-
-        #model = YOLO(hf_hub_download(repo_id="Drazcat-AI/galletas", filename="yolov8_galletas/runs/detect/train/weights/best.pt", local_files_only=True))
-        results = self.model('https://rocketpin-ml-images.s3.amazonaws.com/smu/visits/1120/IMG_20230609_173208_902.jpg', imgsz=800)
-        img = cv2.imread('IMG_20230609_173208_902.jpg')
-        img = cv2.resize(img, (640, 800))
-        
-        for result in results:
-              cv2.imshow("result", result)
-              cv2.waitKey(0)
-     """         
-"""
-class ObjectDetectionHandler:
-    def __init__(self, model_name):
-        self.model_name = model_name
-        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-        self.model = AutoModelForObjectDetection.from_pretrained(self.model_name)
-        self.tokenizer = AutoTokenizer.from_pretrained(self.model_name)
-        self.model.to(self.device)
-
-    def preprocess_image(self, image_path):
-        # Load and preprocess the image
-        image = Image.open(image_path)
-        transform = transforms.Compose([
-            transforms.Resize((224, 224)),  # Resize the image to match the model's input size
-            transforms.ToTensor(),          # Convert the image to a PyTorch tensor
-            transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])  # Normalize the image
-        ])
-        image = transform(image).unsqueeze(0)  # Add batch dimension to the image
-        image = image.to(self.device)
-        return image
-
-    def predict_objects(self, image_path):
-        image = self.preprocess_image(image_path)
-
-        # Perform inference
-        with torch.no_grad():
-            outputs = self.model(image)
-
-        # Process the outputs and extract relevant information
-        # The specifics depend on the model you are using and its output format
-        # Extract the predicted objects and their bounding boxes
-
-        # Sample code to convert the outputs to a dictionary
-        predictions = {
-            "objects": ["object_1", "object_2", ...],
-            "bounding_boxes": [[x_min, y_min, x_max, y_max], ...],
-        }
-
-        return predictions
-
-    def __call__(self, event, context):
-        if "image_path" not in event:
-            return {
-                "statusCode": 400,
-                "body": json.dumps("Error: Please provide an 'image_path' parameter."),
-            }
-
-        image_path = event["image_path"]
-
-        try:
-            predictions = self.predict_objects(image_path)
-            return {
-                "statusCode": 200,
-                "body": json.dumps(predictions),
-            }
-        except Exception as e:
-            return {
-                "statusCode": 500,
-                "body": json.dumps(f"Error: {str(e)}"),
-            }   
-"""            
+            }