from typing import  Dict, List, Any
import torch
import numpy as np
import torch.nn.functional as F

class EndpointHandler():
    def __init__(self, path=""):
        # load the optimized model
        self.model = torch.load(path)




    def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
        """
        Args:
            data (:obj:):
                includes the input data and the parameters for the inference.
        Return:
            A :obj:`list`:. The object returned should be a list of one list like [[{"label": 0.9939950108528137}]] containing :
                - "label": A string representing what the label/class is. There can be multiple labels.
                - "score": A score between 0 and 1 describing how confident the model is for this label/class.
        """
        inputs = data.pop("inputs", data)
        img = inputs["image"]

        # Load the image
        img = np.float32(img) / 255.  # Load and normalize the image

        # Convert to torch tensor and add batch dimension
        img_tensor = torch.from_numpy(img).permute(2, 0, 1).unsqueeze(0)

        # Padding if necessary (to make image dimensions multiples of 4)
        b, c, h, w = img_tensor.shape
        factor = 4  # Assuming factor is 4, based on the code context
        H, W = ((h + factor) // factor) * factor, ((w + factor) // factor) * factor
        padh = H - h if h % factor != 0 else 0
        padw = W - w if w % factor != 0 else 0
        img_tensor = F.pad(img_tensor, (0, padw, 0, padh), 'reflect')

        restored = self.model(img_tensor)
        # postprocess the prediction
        return "OKAY"