import os
import sys

sys.path.append(os.path.join(os.path.dirname(__file__), "..", ".."))

import matplotlib.pyplot as plt
import numpy as np
from PIL import Image

from models.base_model import BaseModelGradCAM
from utils import configs

from .lightning_module import ImageClassificationLightningModule


class DeepLearningGradCAM(BaseModelGradCAM):
    def __init__(
        self,
        name_model: str,
        freeze_model: bool,
        pretrained_model: bool,
        support_set_method: str,
    ):
        super().__init__(name_model, freeze_model, pretrained_model, support_set_method)
        self.init_model()
        self.set_grad_cam()

    def init_model(self):
        self.model = ImageClassificationLightningModule.load_from_checkpoint(
            os.path.join(
                configs.WEIGHTS_PATH,
                self.name_model,
                self.support_set_method,
                "best.ckpt",
            ),
            name_model=self.name_model,
            freeze_model=self.freeze_model,
            pretrained_model=self.pretrained_model,
        )
        self.model = self.model.model
        for layer in self.model.children():
            if hasattr(layer, "reset_parameters") and not self.pretrained_model:
                layer.reset_parameters()
        for param in self.model.parameters():
            param.required_grad = False if not self.freeze_model else True
        self.model.to(self.device)
        self.model.eval()


if __name__ == "__main__":
    model = DeepLearningGradCAM("resnet50", False, True, "5_shot")
    image = np.array(
        Image.open(
            "../../assets/example_images/gon/306e5d35-b301-4299-8022-0c89dc0b7690.png"
        ).convert("RGB")
    )
    gradcam = model.get_grad_cam(image)
    plt.imshow(gradcam)
    plt.show()