import torch
from torch import nn
import torch.optim as optimizer
from torch.autograd import Variable

import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn import datasets
import numpy as np

class simpleNet(nn.Module):
    def __init__(self, in_dim, n_hidden_1, n_hidden_2, out_dim):
        super(simpleNet, self).__init__()
        self.layer1 = nn.Sequential(nn.Linear(in_dim, n_hidden_1), nn.BatchNorm1d(n_hidden_1), nn.ReLU(True))
        self.layer2 = nn.Sequential(nn.Linear(n_hidden_1, n_hidden_2), nn.BatchNorm1d(n_hidden_2), nn.ReLU(True))
        self.layer3 = nn.Sequential(nn.Linear(n_hidden_2, out_dim))
    def forward(self, x):
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        return x



digits = datasets.load_digits()
plt.gray()
plt.matshow(digits.images[0])
plt.show()

print(digits.data.shape)
print(digits.target.shape)
X_train, X_test, y_train, y_test = train_test_split(digits.data, digits.target, test_size=0.3)

learning_rate = 1e-2
model = simpleNet(8 * 8, 100, 50, 10)
criterion = nn.CrossEntropyLoss()
optimizer = optimizer.SGD(model.parameters(), lr=learning_rate)


for epoch in range(2000):
    img = Variable(torch.Tensor(X_train).float(), volatile=True)
    label = Variable(torch.Tensor(y_train).long(), volatile=True)
    out = model(img)
    optimizer.zero_grad()
    loss = criterion(out, label)
    loss.backward()
    optimizer.step()
    print(loss.data)


img = Variable(torch.Tensor(X_test).float(), volatile=True)
out = model(img)

result = []
for i in range(0, len(out.data.numpy())):
    result.append(np.argmax(out[i].data.numpy()))
print(result)
print(y_test)

sum = 0
for i in range(0, len(y_test)):
    if result[i] == y_test[i]:
        sum += 1
print(sum / len(result))