-
Notifications
You must be signed in to change notification settings - Fork 0
/
rnn_predictor.py
175 lines (147 loc) · 6.73 KB
/
rnn_predictor.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
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
# -*- coding: utf-8 -*-
from __future__ import division, print_function, absolute_import
import numpy as np
import matplotlib.pyplot as plt
import tensorflow as tf
import time
import os
from signal_provider import SignalProvider
from hyper_parameter import HyperParameter
from models import dynamic_rnn_model, seq2seq_model, position_rnn_model, position_seq2seq_model
class SignalPredictor(object):
def __init__(self, params):
self._batch_size = params.rnn_batch_size
self._input_steps = params.rnn_input_steps
self._predict_steps = params.rnn_predict_steps
self._input_depth = params.rnn_input_depth
self._predict_depth = params.rnn_predict_depth
self._hidden = params.rnn_hidden
self._train_epoch = params.rnn_train_epoch
self._model_dir = params.rnn_model_dir
if not os.path.exists(self._model_dir):
os.mkdir(self._model_dir)
self._data_provider = SignalProvider(self._batch_size,
input_steps=self._input_steps, predict_steps=self._predict_steps)
self._input_X = tf.placeholder(tf.float32, [self._batch_size, self._input_steps, self._input_depth])
self._truth_Y = tf.placeholder(tf.float32, [self._batch_size, self._predict_steps, self._predict_depth])
self._X = tf.placeholder(tf.float32, [None, self._input_steps, self._input_depth])
global_step = tf.train.get_or_create_global_step()
inc_step = tf.assign_add(global_step, 1)
with tf.name_scope("train"):
with tf.variable_scope("spnn", reuse=None):
self._trained_Y = dynamic_rnn_model(self._input_X, True, params)
# self._trained_Y, self._reg_loss = position_seq2seq_model(self._input_X, True, params)
# self._trained_Y, self._reg_loss = seq2seq_model(self._input_X, True, params)
with tf.name_scope("eval"):
with tf.variable_scope("spnn", reuse=True):
self._predict = dynamic_rnn_model(self._X, False, params)
# self._predict, _ = position_seq2seq_model(self._X, False, params)
# self._predict, _ = seq2seq_model(self._X, False, params)
self._loss = self._get_loss()
self._train_step = self._get_train_step()
self._sess = tf.InteractiveSession()
def _get_loss(self):
# loss = tf.losses.absolute_difference(self._trained_Y, self._truth_Y)
loss = tf.losses.mean_squared_error(self._trained_Y, self._truth_Y)
return loss
def _get_train_step(self):
# train_step = tf.train.AdamOptimizer().minimize(self._loss + self._reg_loss)
train_step = tf.train.AdamOptimizer().minimize(self._loss)
return train_step
def train(self):
saver = tf.train.Saver()
# start training
tf.global_variables_initializer().run()
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess=self._sess, coord=coord)
time_s = time.clock()
epoch = 0
i = 0
loss_avg = 0
while epoch < self._train_epoch:
x, y, nid = self._data_provider.get_next_batch()
y = y[:,:,:self._predict_depth]
p, loss, _ = self._sess.run([self._trained_Y, self._loss, self._train_step],
feed_dict={self._input_X: x, self._truth_Y: y})
loss_avg += loss
i += 1
if nid == 0:
epoch += 1
print("[INFO] after {} training steps, loss is {}, "
"time elapse {}".format(epoch, loss_avg / i, time.clock() - time_s))
i = 0
loss_avg = 0
time_s = time.clock()
saver.save(self._sess, os.path.join(self._model_dir, "spnn_si{}so{}dp{}.ckpt".format(
self._input_steps, self._predict_steps, self._hidden
)))
coord.request_stop()
coord.join(threads)
def load_model(self):
model_fn = os.path.join(self._model_dir, "spnn_si{}so{}dp{}.ckpt".format(
self._input_steps, self._predict_steps, self._hidden
))
saver = tf.train.Saver()
saver.restore(self._sess, model_fn)
def evaluate_pn(self):
init_data = self._data_provider.evaluate_dat()
init_len = init_data.shape[1]
assert init_len >= self._input_steps
predicted = init_data
predicted_len = self._data_provider.evaluate_length()
while len(predicted[0]) < init_len + predicted_len:
x = predicted[:, len(predicted[0]) - self._input_steps:]
x = (x - self._data_provider.norm_mean) / self._data_provider.norm_std
y = self._sess.run(self._predict,
feed_dict={self._X: x})
y = y * self._data_provider.norm_std[:self._predict_depth] + \
self._data_provider.norm_mean[:self._predict_depth]
predicted = np.concatenate((predicted, y), axis=1)
predicted = predicted[:, init_len: init_len + predicted_len]
return self._data_provider.evaluate(predicted)
def evaluate_p1(self):
init_data = self._data_provider.evaluate_dat_v2()
init_len = init_data.shape[2]
assert init_len >= self._input_steps
predicted = []
for i in range(init_data.shape[1]):
x = init_data[:, i, init_len - self._input_steps:]
x = (x - self._data_provider.norm_mean) / self._data_provider.norm_std
y = self._sess.run(self._predict,
feed_dict={self._X: x})
y = y[:, 0] * self._data_provider.norm_std[:self._predict_depth] + \
self._data_provider.norm_mean[:self._predict_depth]
predicted.append(y)
predicted = np.stack(predicted, axis=1)
return self._data_provider.evaluate(predicted)
def evaluate_once(param1):
tf.reset_default_graph()
with tf.device('/cpu:0'):
params = HyperParameter()
params.rnn_predict_steps = param1
sp = SignalPredictor(params)
sp.load_model()
rmse = sp.evaluate_pn()
return rmse
def train_once(param1, param2):
tf.reset_default_graph()
with tf.device('/cpu:0'):
params = HyperParameter()
params.rnn_hidden = param2
params.rnn_input_steps = param1
sp = SignalPredictor(params)
sp.train()
rmse = sp.evaluate_p1()
return rmse
if __name__ == '__main__':
params1 = [5]
params2 = [20]
res = np.zeros((len(params1), len(params2)), dtype=np.float32)
p = 0
for i in range(len(params1)):
for j in range(len(params2)):
rmse = train_once(params1[i], params2[j])
# rmse = evaluate_once(params[i])
res[i, j] = rmse
print(res)
print(res)