from torchvision import datasets, transforms
from torch.utils.data import DataLoader, Dataset
import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
import os
class CNN(nn.Module):
    def __init__(self):
        super().__init__()
        self.Sq1 = nn.Sequential(         
            nn.Conv2d(in_channels=1, out_channels=16, kernel_size=5, stride=1, padding=2),   # (16, 28, 28)                           #  output: (16, 28, 28)
            nn.ReLU(),                    
            nn.MaxPool2d(kernel_size=2),    # (16, 14, 14)
        )
        self.Sq2 = nn.Sequential(
            nn.Conv2d(in_channels=16, out_channels=32, kernel_size=5, stride=1, padding=2),  # (32, 14, 14)
            nn.ReLU(),                      
            nn.MaxPool2d(2),                # (32, 7, 7)
        )
        self.out = nn.Linear(32 * 7 * 7, 100)  
    def forward(self, x):
        x = self.Sq1(x)
        x = self.Sq2(x)
        x = x.view(x.size(0), -1)    
        x = self.out(x)
        ## Sigmoid activation   
        output = F.sigmoid(x)  # 1/(1+e**(-x))
        return output
def loss_fn(pred, target):
    return -(target * torch.log(pred) + (1 - target) * torch.log(1 - pred)).sum()
def multilabel_generate(label):
    Y1 = F.one_hot(label, num_classes = 100)
    Y2 = F.one_hot(label+10, num_classes = 100)
    Y3 = F.one_hot(label+50, num_classes = 100) 	
    multilabel = Y1+Y2+Y3
    return multilabel
        
# def multilabel_generate(label):
# 	multilabel_dict = {}
# 	multi_list = []
# 	for i in range(label.shape[0]):
# 		multi_list.append(multilabel_dict[label[i].item()])
# 	multilabel_tensor = torch.tensor(multi_list)
#     return multilabel
def train():
    epoches = 10
    mnist_net = CNN()
    mnist_net.train()
    opitimizer = optim.SGD(mnist_net.parameters(), lr=0.002)
    mnist_train = datasets.MNIST("mnist-data", train=True, download=True, transform=transforms.ToTensor())
    train_loader = torch.utils.data.DataLoader(mnist_train, batch_size= 128, shuffle=True)
    for epoch in range(epoches):
    	loss = 0 
    	for batch_X, batch_Y in train_loader:
    		opitimizer.zero_grad()
    		outputs = mnist_net(batch_X)
    		loss = loss_fn(outputs, multilabel_generate(batch_Y)) / batch_X.shape[0]
    		loss.backward()
    		opitimizer.step()
    		print(loss)
if __name__ == '__main__':
	train()