import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.utils.data import DataLoader
import numpy as np
from sklearn.metrics import classification_report
from architectures.Model import Model

class WF_Module(nn.Module):
    """
    Pure PyTorch Neural Network module graph.
    Keeps parameter registration and autograd tracking separate from 
    the framework's high-level Model abstractions to prevent recursion collisions.
    """
    def __init__(self, original_model: nn.Module, num_classes: int):
        super().__init__()
    
        self.original_model = original_model
        
        # Target layer for weight filtering (layer4 block 1 conv2 or conv3 depending on arch)
        last_layer = original_model.layer4[1]

        # Some versions are limited to 2 convolutional layers
        if hasattr(last_layer, "conv3"):
            self.target_conv = last_layer.conv3
        else:
            self.target_conv = last_layer.conv2

        # Completely freeze the original ResNet parameters
        for param in self.parameters():
            param.requires_grad = False
            
        # Initialize the alpha parameter matrix (Rows = Classes, Cols = Channels)
        out_channels = self.target_conv.weight.shape[0]
        self.alpha = nn.Parameter(torch.full((num_classes, out_channels), 3.0))

    '''
    Poppi et_al's Single-shot multiclass unlearning.
    This calculation happens only once to generate the mask. once the mask is generated,
    Unlearning and remembering becomes a matter of switching gates on and off.
    '''
    def forward(self, x: torch.Tensor, target_class_indices: torch.Tensor) -> torch.Tensor:
        # we linearly loop through layers 1 to 4[block 1] (for ResNet) 
        # for i in M_{|L|} do  l <- l[i]
        x = self.original_model.maxpool(self.original_model.relu(self.original_model.bn1(self.original_model.conv1(x))))
        x = self.original_model.layer1(x)
        x = self.original_model.layer2(x)
        x = self.original_model.layer3(x)
        x = self.original_model.layer4[0](x)
        
        # The second block execute its internal transformations natively
        # This handles conv1->conv2 (ResNet18) or conv1->conv2->conv3 (ResNet50) automatically!
        # Xi+1 <- l(Xi, ˆwl)
        x = self.original_model.layer4[1](x)
        
        # Apply mask dynamically to the completed block feature map
        # wl <- αl[Yunl] ⊙ ˆwl
        batch_alpha = self.alpha[target_class_indices]
        mask = torch.sigmoid(batch_alpha).view(x.size(0), -1, 1, 1)
        x = x * mask
        
        # Remaining standard head steps
        x = self.original_model.avgpool(x)
        x = torch.flatten(x, 1)
        # so here we are returning the output logits
        # the result of classification is then 
        # argmax(x)
        return self.original_model.fc(x)
    


class WF_Net_Model(Model):
    def __init__(self, device, size, original_model: nn.Module, target_class_index: int):
        self.device = device
        self.size = size
        self.wf_module = WF_Module(
            original_model = original_model, 
            num_classes = size
        ).to(self.device)
        
        # this index indicates which row of the mask should be active (gate closed).
        self.target_class_index = target_class_index
        self.model = self.wf_module

    def get(self):
        return self.wf_module
    
    '''
    We override the evaluate method from the base class, 
    because how we evaluate is different here from that of a normal torch nn.Module object

    '''
    def evaluate(self, loader, mode="eval"):

        self.wf_module.eval()
        all_preds, all_labels = [], []
        print(f"\nEvaluating Domain: [{mode}]...")
        
        with torch.no_grad():
            for inputs, labels in loader:
                inputs, labels = inputs.to(self.device), labels.to(self.device)
                
                # we apply the filter
                gate_signals = torch.full((inputs.size(0),), self.target_class_index, dtype=torch.long, device=self.device)
                
                # pass prediction through the filter
                outputs = self.wf_module(inputs, target_class_indices=gate_signals)

                # return argmax(x)    
                _, predicted = torch.max(outputs, 1)
                all_preds.extend(predicted.cpu().numpy())
                all_labels.extend(labels.cpu().numpy())

        classes = sorted(list(set(all_labels)))
        accuracy = 100 * (np.array(all_preds) == np.array(all_labels)).sum() / len(all_labels)
        
        print(f"Test Accuracy: {accuracy:.2f}%")
        print(classification_report(all_labels, all_preds, labels=classes, zero_division=0))
        report = classification_report(all_labels, all_preds, labels=classes, output_dict=True, zero_division=0)
        
        return accuracy, report
    
    def eval(self):
        """Safely intercept any fallback base class calls targeting .eval()"""
        self.wf_module.eval()