-
Notifications
You must be signed in to change notification settings - Fork 0
/
lstm.py
147 lines (122 loc) · 4.96 KB
/
lstm.py
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
143
144
145
146
147
import tensorflow as tf
import pandas as pd
import math
import random
from sklearn import preprocessing
from sklearn.utils import shuffle
import numpy as np
np.random.seed(2)
random.seed(2)
data = pd.read_csv("./datasets/combined_data_set3.csv", header=None, skiprows=[0], usecols=(range(1,33)))
#data2 = pd.read_csv("HomeB.csv", header=None, skiprows=[0], usecols=(range(17, 34)))
#frames = [data1, data2]
#data = pd.concat(frames, axis=1)
dataset = data.values
num_features = 32
num_hidden = 256
num_classes = 1
batch_size = 16
time_steps = 10
train_size = int(dataset.shape[0] * 0.70)
# Make dataset random from everywhere
# rows = np.arange(dataset.shape[0] - time_steps)
# np.random.shuffle(rows)
# train_rows = rows[:train_size]
# te_rows = rows[train_size:]
#
# train = dataset[train_rows, :]
# scalar = preprocessing.StandardScaler().fit(train)
# sample test dataset from the end
train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]
scalar = preprocessing.StandardScaler().fit(train)
trainX = scalar.transform(train)
trainY = dataset[1:len(trainX) + 1, 11]
testX = scalar.transform(test)
testY = dataset[len(trainX) + 1:, 11]
# train = np.column_stack((trainX, trainY))[:-3, :]
train = np.column_stack((trainX, trainY))[:-9, :]
train = train.reshape(-1, time_steps, 33)
testX = testX[:-1] # make test y same dimesnsions as test
test = np.column_stack((testX, testY))[:-7, :]
test = test.reshape(-1, time_steps, 33)
# np.random.shuffle(test)
np.random.shuffle(train)
x_ph = tf.placeholder('float', [None, time_steps - 1, num_features], name='x')
y_ph = tf.placeholder('float', [None, num_classes], name='y')
dropout_ph = tf.placeholder('float', [], name='dropout')
def next_batch(i):
x_batch = train[i:i + batch_size, :time_steps - 1, 0:num_features]
y_batch = train[i:i + batch_size, time_steps - 1, num_features]
return x_batch, y_batch
def te_batch():
x_batch = test[:, :time_steps - 1, 0:num_features]
y_batch = test[:, time_steps - 1, num_features]
return x_batch, y_batch.reshape(-1, 1)
def lst_model(x):
x = tf.unstack(x, time_steps - 1, 1)
rnn_cell = tf.contrib.rnn.MultiRNNCell(
[
# tf.contrib.rnn.DropoutWrapper(
# tf.contrib.rnn.BasicLSTMCell(num_hidden, activation='relu'),
# input_keep_prob=dropout_ph,
# output_keep_prob=dropout_ph,
# state_keep_prob=dropout_ph
# ),
# tf.contrib.rnn.DropoutWrapper(
tf.contrib.rnn.BasicLSTMCell(num_hidden, activation='relu'),
# input_keep_prob = dropout_ph,
# output_keep_prob = dropout_ph,
# state_keep_prob = dropout_ph
# )
]
)
# lstm_cell = tf.contrib.rnn.BasicLSTMCell(num_hidden, forget_bias=1.0)
outputs, states = tf.nn.static_rnn(rnn_cell, x, dtype=tf.float32)
return tf.layers.dense(outputs[-1], units=1)
def accuracy(x, y, data_size):
acc = 0
for i in range(0, data_size):
if abs(x[i] - y[i]) < 5.5:
acc += 1
return float(acc / data_size)
prediction = lst_model(x_ph)
cost = tf.losses.huber_loss(y_ph, prediction)
optimizer = tf.train.AdamOptimizer(learning_rate=0.001).minimize(cost)
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
training_steps = 1000
acc_arr_te = []
acc_arr_tr = []
loss_arr_te = []
loss_arr_tr = []
for epoch in range(0, training_steps):
np.random.shuffle(train)
for i in range(int(len(train) - batch_size)):
batch_x, batch_y = next_batch(i)
batch_x = batch_x.reshape((batch_size, time_steps - 1, num_features))
batch_y = batch_y.reshape(-1, 1)
sess.run(optimizer, feed_dict={x_ph: batch_x, y_ph: batch_y})
c, prd = sess.run([cost, prediction], feed_dict={x_ph: batch_x, y_ph: batch_y})
te_x, te_y = te_batch()
loss_te = cost.eval({x_ph: te_x, y_ph: te_y})
print('===================================================')
print('Epoch: {}'.format(epoch))
print('Huber Loss on Train:', c)
print('Huber Loss on Test:', loss_te)
# print("Prediction eval:", prediction.eval({x_ph: te_x}))
acc_te = accuracy(prediction.eval({x_ph: te_x}), te_y, len(te_y))
acc_tr = accuracy(prd, batch_y, len(batch_y))
acc_arr_te.append([epoch, acc_te])
acc_arr_tr.append(([epoch, acc_tr]))
loss_arr_tr.append([epoch, c])
loss_arr_te.append([epoch, loss_te])
print("Accuracy on Train:", float(acc_tr))
print("Accuracy on Test:", float(acc_te))
acc_arr_te = np.array(acc_arr_te)
acc_arr_tr = np.array(acc_arr_tr)
loss_arr_te = np.array(loss_arr_te)
loss_arr_tr = np.array(loss_arr_tr)
np.save("./LSTM/threshold0-2-st10-lr0001-batch16_te_acc.npy", acc_arr_te)
np.save("./LSTM/threshold0-2-st10-lr0001-batch16_tr_acc.npy", acc_arr_tr)
np.save("./LSTM/threshold0-2-st10-lr0001-batch16_te_loss.npy", loss_arr_te)
np.save("./LSTM/threshold0-2-st10-lr0001-batch16_tr_loss.npy", loss_arr_tr)