-
Notifications
You must be signed in to change notification settings - Fork 6
/
cifar.py
180 lines (159 loc) · 8.14 KB
/
cifar.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
176
177
178
179
180
from __future__ import (absolute_import, division, print_function, unicode_literals)
import pickle
import matplotlib.pyplot as plt
import numpy as np
import pydot
import tensorflow as tf
import tqdm
from keras.utils import vis_utils
from tensorflow.keras import datasets, layers, models
vis_utils.pydot = pydot
def unpickle(file):
with open(file, 'rb') as fo:
dict = pickle.load(fo, encoding='bytes')
return dict
test, cross = unpickle('./test_batch'), unpickle(f'./data_batch_5')
data, labels = [], []
for i in range(5):
cifar = unpickle(f'./data_batch_{i + 1}')
if i == 0:
data = cifar[b'data'] / 255
labels = np.array(cifar[b'labels'])
else:
data = np.append(data, cifar[b'data'] / 255, axis=0)
labels = np.append(labels, np.array(cifar[b'labels']), axis=0)
def network(data, labels, test, cross):
data.resize((data.shape[0], 1, data.shape[-1]))
cross[b'data'].resize((cross[b'data'].shape[0], 1, cross[b'data'].shape[-1]))
test[b'data'].resize((test[b'data'].shape[0], 1, test[b'data'].shape[-1]))
model = models.Sequential([
layers.Dense(512, activation='relu'),
layers.Dense(256, activation='relu'),
layers.Dense(128, activation='relu'),
layers.Dense(10, activation='softmax')
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
history = model.fit(data, labels, epochs=10, validation_data=(cross[b'data'] / 255, np.array(cross[b'labels'])))
plt.plot(history.history['accuracy'], label='accuracy')
plt.plot(history.history['val_accuracy'], label='val_accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.legend(loc='lower right')
plt.savefig('nn_evaluation.png', dpi=600)
cross_loss, cross_acc = model.evaluate(cross[b'data'] / 255, np.array(cross[b'labels']), verbose=2)
model.save(f'nn{cross_acc:.2}.h5')
vis_utils.plot_model(model, to_file='nn.png', show_shapes=True, show_layer_names=True, expand_nested=True, dpi=600)
print(f'Cross Validation Accuracy: {cross_acc}, Cross Validation lost: {cross_loss}')
test_loss, test_acc = model.evaluate(test[b'data'] / 255, np.array(test[b'labels']), verbose=2)
print(f'Test accuracy: {test_acc}, Test lost: {test_loss}')
print(model.summary())
def CNN(data, labels, test, cross):
data = np.array([i.reshape((3, 1024)).T.reshape(32, 32, 3) for i in data])
cross[b'data'] = np.array([i.reshape((3, 1024)).T.reshape(32, 32, 3) for i in cross[b'data']])
test[b'data'] = np.array([i.reshape((3, 1024)).T.reshape(32, 32, 3) for i in test[b'data']])
model = models.Sequential([
layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)),
layers.MaxPooling2D((2, 2)),
layers.Conv2D(64, (3, 3), activation='relu'),
layers.MaxPooling2D((2, 2)),
layers.Conv2D(64, (3, 3), activation='relu'),
layers.Flatten(),
layers.Dense(64, activation='relu'),
layers.Dense(10, activation='softmax')
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
history = model.fit(data, labels, epochs=10, validation_data=(cross[b'data'] / 255, np.array(cross[b'labels'])))
plt.plot(history.history['accuracy'], label='accuracy')
plt.plot(history.history['val_accuracy'], label='val_accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.ylim([0.5, 1])
plt.legend(loc='lower right')
plt.savefig('cnn_evaluation.png', dpi=600)
cross_loss, cross_acc = model.evaluate(cross[b'data'] / 255, np.array(cross[b'labels']), verbose=2)
model.save(f'cnn{cross_acc:.2}.h5')
from tensorflow.keras.utils import plot_model
plot_model(model, to_file='cnn.png', show_shapes=True, show_layer_names=True, expand_nested=True, dpi=600)
print(f'Cross Validation Accuracy: {cross_acc}, Cross Validation lost: {cross_loss}')
test_loss, test_acc = model.evaluate(test[b'data'] / 255, np.array(test[b'labels']), verbose=2)
print(f'Test accuracy: {test_acc}, Test lost: {test_loss}')
print(model.summary())
class NearestNeighbor:
def __init__(self, data, labels, test, cross):
self.test = test
self.data = data
self.labels = labels
self.test[b'data'] = test[b'data']
self.test[b'labels'] = test[b'labels']
self.cross = cross
self.train()
def train(self):
predict = self.predict()
accuracy = np.mean(predict == self.test[b'labels'])
print(f'Accuracy:\t{accuracy}')
def predict(self, k=7):
predict = np.zeros(self.test[b'data'].shape[0], dtype=self.labels.dtype)
for i in tqdm.tqdm(range(self.test[b'data'].shape[0])):
L1 = np.sum(np.abs(self.data - self.test[b'data'][i, :]), axis=1)
closest = self.labels[np.argsort(L1)[:k]]
unique, indices = np.unique(closest, return_inverse=True)
predict[i] = unique[np.argmax(np.bincount(indices))]
return predict
def rnn(data, labels, test, cross, first_exec=True):
import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()
size = 32 # 32 * 32
timesteps = 32
hidden_layer = 256
classes = 10
params = {"learning_rate": 0.001, "training_iters": 10000, "batch_size": 64}
test_data, test_labels = test[b'data'] / 255, test[b'labels']
# 将RGB值转为灰度值
print('Converting data......')
data_array = np.array([[[item[index], item[index + 1024], item[index + 1024 * 2]] for index in range(1024)]
for item in tqdm.tqdm(data)])
test_array = np.array([[[item[index], item[index + 1024], item[index + 1024 * 2]] for index in range(1024)]
for item in tqdm.tqdm(test_data)])
data = np.array([[data_array[i, j].dot([0.299, 0.587, 0.114]) for j in range(data_array.shape[1])]
for i in tqdm.tqdm(range(data_array.shape[0]))])
test = np.array([[test_array[i, j].dot([0.299, 0.587, 0.114]) for j in range(test_array.shape[1])]
for i in tqdm.tqdm(range(test_array.shape[0]))])
labels = np.array([[1 if i == row else 0 for i in range(10)] for row in tqdm.tqdm(labels)])
test_labels = np.array([[1 if i == row else 0 for i in range(10)] for row in tqdm.tqdm(test_labels)])
# 按照 tutorial 定义 RNN 模型
x = tf.placeholder("float", [None, timesteps, size])
y = tf.placeholder("float", [None, classes])
weights = tf.Variable(tf.random_normal([hidden_layer, classes]), name='weights')
biases = tf.Variable(tf.random_normal([classes]), name='biases')
def rnn_model(x, weights, biases):
x = tf.transpose(x, [1, 0, 2])
x = tf.reshape(x, [-1, size])
x = tf.split(x, timesteps, axis=0)
lstm_cell = tf.nn.rnn_cell.LSTMCell(hidden_layer, forget_bias=1.0)
outputs, states = tf.nn.static_rnn(lstm_cell, x, dtype=tf.float32)
return tf.matmul(outputs[-1], weights) + biases
pred = rnn_model(x, weights, biases)
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(logits=pred, labels=y))
optimizer = tf.train.AdamOptimizer(learning_rate=params['learning_rate']).minimize(cost)
correct_pred = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))
# 训练模型
print('Training......')
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
# 一轮一轮地训练模型
for step in tqdm.tqdm(range(1, int(params['training_iters'] / params['batch_size']) + 1)):
batch_x = data[(step - 1) * params['batch_size']:step * params['batch_size']].reshape(
(params['batch_size'], timesteps, size))
batch_y = labels[(step - 1) * params['batch_size']:step * params['batch_size']]
sess.run(optimizer, feed_dict={x: batch_x, y: batch_y})
# 测试评估模型
print("Accuracy:",
sess.run(accuracy, feed_dict={
x: test[:128].reshape((-1, timesteps, size)),
y: test_labels[:128]
}))
network(data, labels, test, cross)
CNN(data, labels, test, cross)
NearestNeighbor(data, labels, test, cross)
rnn(data, labels, test, cross, 0)