DBN-and-RBM-in-pytorch/mnist_tutorial_DBN.py at master · AlvinLXS/DBN-and-RBM-in-pytorch · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142

# coding: utf-8

# In[1]:


# Importing the libraries
from DBN import DBN
import torch
import torchvision
from torchvision import datasets,transforms
from torch.utils.data import Dataset,DataLoader

import matplotlib
import matplotlib.pyplot as plt

import math
import numpy as np


# In[2]:


#Loading MNIST dataset
mnist_data = datasets.MNIST('../data', train=True, download=True,
                    transform=transforms.Compose(
                    [transforms.ToTensor(),transforms.Normalize((0.1307,), (0.3081,))]))


# In[3]:


mnist_data.train_data = (mnist_data.train_data.type(torch.cuda.FloatTensor)/255).bernoulli()


# In[4]:


#Lets us visualize a number from the data set
idx = 5
img = mnist_data.train_data[idx]
print("The number shown is the number: {}".format(mnist_data.train_labels[idx]) )
plt.imshow(img , cmap = 'gray')
plt.show()


# In[5]:


# I have have set these hyper parameters although you can experiment with them to find better hyperparameters.
dbn_mnist = DBN(visible_units=28*28 ,
                hidden_units=[23*23 ,18*18] ,
                k = 5,
                learning_rate = 0.01,
                learning_rate_decay = True,
                xavier_init = True,
                increase_to_cd_k = False,
                use_gpu = True)


# In[6]:


num_epochs = 2
batch_size = 1000
tensor_x = mnist_data.train_data.cuda()
tensor_y = mnist_data.train_labels.cuda()
dbn_mnist.train_static(tensor_x,tensor_y,num_epochs , batch_size)


# In[ ]:


# visualising layer 1
# learned_weights = dbn_mnist.rbm_layers[0].W.transpose(0,1).numpy()
# plt.show()
# fig = plt.figure(3, figsize=(10,10))
# for i in range(25):
#     sub = fig.add_subplot(5, 5, i+1)
#     sub.imshow(learned_weights[i,:].reshape((28,28)), cmap=plt.cm.gray)
# plt.show()


# In[ ]:


# visualising layer 2
# learned_weights = dbn_mnist.rbm_layers[1].W.transpose(0,1).numpy()
# plt.show()
# fig = plt.figure(3, figsize=(10,10))
# for i in range(25):
#     sub = fig.add_subplot(5, 5, i+1)
#     sub.imshow(learned_weights[i,:].reshape((23,23)), cmap=plt.cm.gray)
# plt.show()


# In[ ]:


number = 5 #A number between 0 and 10.

particular_mnist = []

limit = mnist_data.train_data.shape[0]
# limit = 60000
for i in range(limit):
    if(mnist_data.train_labels[i] == number):
        particular_mnist.append(mnist_data.train_data[i].cpu().numpy())
# particular_mnist = np.array(particular_mnist)
len(particular_mnist)
# mnist_data.train_data


# In[ ]:


train_data = torch.stack([torch.Tensor(i) for i in particular_mnist])
train_label = torch.stack([torch.Tensor(number) for i in range(len(particular_mnist))])


# In[ ]:


dbn_mnist.train_static(train_data,train_label,num_epochs , batch_size)


# In[ ]:


idx = 3
img = mnist_data.train_data[idx]
reconstructed_img = img.view(1,-1).type(torch.FloatTensor)

_,reconstructed_img= dbn_mnist.reconstruct(reconstructed_img)

reconstructed_img = reconstructed_img.view((28,28))
print("The original number: {}".format(mnist_data.train_labels[idx]))
plt.imshow(img , cmap = 'gray')
plt.show()
print("The reconstructed image")
plt.imshow(reconstructed_img , cmap = 'gray')
plt.show()