from __future__ import print_function
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import torchvision
from torchvision import datasets, transforms
import torchvision.models as models
import matplotlib.pyplot as plt
import numpy as np


# class Net(nn.Module):
#     def __init__(self):
#         super(Net, self).__init__()
#         self.conv1 = nn.Conv2d(1, 64, kernel_size=5)
#         self.conv2 = nn.Conv2d(64, 32, kernel_size=3)
#         self.conv3 = nn.Conv2d(32, 16, kernel_size=5)
#         self.fc1 = nn.Linear(1*16, 10)
#
#     def forward(self, x):
#
#         x = F.relu(F.max_pool2d(self.conv1(x), 2))
#         x = F.relu(F.max_pool2d(self.conv2(x), 2))
#         x = F.relu(self.conv3(x), 2)
#
#         x = x.view(-1, 1*16)
#         x = F.relu(self.fc1(x))
#
#         return F.log_softmax(x, dim=1)

# class Net(nn.Module):
#     def __init__(self):
#         super(Net, self).__init__()
#         self.conv1 = nn.Conv2d(1,  16, kernel_size=5)
#         self.conv2 = nn.Conv2d(16, 16, kernel_size=3)
#         self.conv3 = nn.Conv2d(16, 10, kernel_size=5)
#
#     def forward(self, x):
#
#         x = F.sigmoid(F.max_pool2d(self.conv1(x), 2))
#         x = F.sigmoid(F.max_pool2d(self.conv2(x), 2))
#         x = self.conv3(x)
#
#         x = x.view(-1, 1*10)
#
#         return F.log_softmax(x, dim=1)


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        layers = []
        layers += [nn.Conv2d(1,  8, kernel_size=5),nn.MaxPool2d(kernel_size=2, stride=2),nn.Sigmoid()]
        layers += [nn.Conv2d(8,  8, kernel_size=3),nn.MaxPool2d(kernel_size=2, stride=2),nn.Sigmoid()]
        layers += [nn.Conv2d(8,  10, kernel_size=5)]
        self.features = nn.Sequential(*layers)

        # self.conv1 = nn.Conv2d(1,  8, kernel_size=5)
        # self.conv2 = self.__conv(8,  8, kernel_size=3)
        # self.conv3 = self.__conv(8, 10, kernel_size=5)

    def forward(self, x):


        # x = F.sigmoid(F.max_pool2d(self.conv1(x), 2))
        # x = F.sigmoid(F.max_pool2d(self.conv2(x), 2))
        # x = self.conv3(x)
        x = self.features(x)
        x = x.view(-1, 1*10)

        return F.log_softmax(x, dim=1)