DropNeuron/alexnet.py at master · panweihit/DropNeuron · 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
143
144
145
146
147
148
149
150
151
152
153
154
# -*- coding: utf-8 -*-

""" AlexNet.

Applying 'Alexnet' to Oxford's 17 Category Flower Dataset classification task.

References:
    - Alex Krizhevsky, Ilya Sutskever & Geoffrey E. Hinton. ImageNet
    Classification with Deep Convolutional Neural Networks. NIPS, 2012.
    - 17 Category Flower Dataset. Maria-Elena Nilsback and Andrew Zisserman.

Links:
    - [AlexNet Paper](http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf)
    - [Flower Dataset (17)](http://www.robots.ox.ac.uk/~vgg/data/flowers/17/)

================================How to run this script=================================
The specification of Alexnet using TensorFlow is complicated,
therefore we are using modular framework to make the modelling simpler but transparent

In this script, please install 'TFLearn' available at https://github.com/tflearn/tflearn

This is a modification of the example in TFLearn, available at:
https://github.com/tflearn/tflearn/blob/master/examples/images/alexnet.py

You can run the following command without DropNeuron or Dropout

   $ python alexnet.py 0 0 0.0005 1 0.001

Author: Wei Pan
Contact: w.pan11@imperial.ac.uk
         dropneuron@gmail.com
"""

from __future__ import division, print_function, absolute_import

import tensorflow as tf
from regularizers import *

import sys
import numpy as np
import tflearn
from tflearn.layers.core import input_data, dropout, fully_connected
from tflearn.layers.conv import conv_2d, max_pool_2d
from tflearn.layers.normalization import local_response_normalization
from scipy.io import savemat

import tflearn.datasets.oxflower17 as oxflower17
# import oxflower17 as oxflower17
trainX, trainY = oxflower17.load_data(one_hot=True)

n_class = 17

learning_rate_ini = 0.001
lambda_l1 = float(sys.argv[1])
lambda_l2 = float(sys.argv[2])
lambda_dropneuron = float(sys.argv[3])
keep_prob = float(sys.argv[4])   # keep_prob \in (0, 1]
threshold = float(sys.argv[5])


def prune(x):
    # Due to machine precision, there is no absolute zeros in the solution usually;
    # Therefore, we set a very small threshold to prune some parameters:
    # However, the test error is obtained after pruning
    y = trainer.get_weights(x)
    y = np.asarray(y)
    low_values_indices = abs(y) < threshold
    y[low_values_indices] = 0
    return y

# Define a dnn using Tensorflow
with tf.Graph().as_default():
    # Building 'AlexNet'
    # Placeholders for data and labels
    X = tf.placeholder(shape=[None, 224, 224, 3], dtype=tf.float32)
    Y = tf.placeholder(shape=(None, n_class), dtype=tf.float32)
    # network = input_data(shape=[None, 224, 224, 3],  name='input')
    network = X
    network = conv_2d(network, 96, 11, strides=4, activation='relu')
    network = max_pool_2d(network, 3, strides=2)
    network = local_response_normalization(network)
    network = conv_2d(network, 256, 5, activation='relu')
    network = max_pool_2d(network, 3, strides=2)
    network = local_response_normalization(network)
    network = conv_2d(network, 384, 3, activation='relu')
    network = conv_2d(network, 384, 3, activation='relu')
    network = conv_2d(network, 256, 3, activation='relu')
    network = max_pool_2d(network, 3, strides=2)
    network = local_response_normalization(network)
    network = fully_connected(network, 4096, activation='tanh')
    network = dropout(network, keep_prob)
    # network = fully_connected(network, 4096, activation='tanh')
    # network = dropout(network, 0.5)
    # network = fully_connected(network, n_class, activation='softmax')
    network_fc1 = fully_connected(network, 4096, activation='tanh')
    network = dropout(network_fc1, keep_prob)
    network_fc2 = fully_connected(network, n_class, activation='softmax')
    network = network_fc2
    fc1_weights = network_fc1.W
    fc2_weights = network_fc2.W

    loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(network, Y))

    regularizers = tf.reduce_mean(li_regularizer(.1)(fc1_weights))
    regularizers += tf.reduce_mean(lo_regularizer(.1)(fc1_weights))
    regularizers += tf.reduce_mean(li_regularizer(.1)(fc2_weights))
    regularizers += tf.reduce_mean(lo_regularizer(.1)(fc2_weights))
    loss += lambda_dropneuron * regularizers


    # optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.1)
    optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate_ini)
    # optimizer = tf.train.MomentumOptimizer(learning_rate, 0.9)

    accuracy = tf.reduce_mean(
        tf.cast(tf.equal(tf.argmax(network, 1), tf.argmax(Y, 1)), tf.float32),
        name="acc")

    # Define a train op
    trainop = tflearn.TrainOp(loss=loss, optimizer=optimizer,
                              metric=accuracy, batch_size=128)

    # Tensorboard logs stored in /tmp/tflearn_logs/. Using verbose level 2.
    trainer = tflearn.Trainer(train_ops=trainop,
                            tensorboard_dir='tflearn_logs/',
                            tensorboard_verbose=2)

    # Training
    # trainer.fit({X: trainX, Y: trainY}, val_feed_dicts={X: testX, Y: testY},
    #             n_epoch=10, show_metric=True, run_id='Summaries_example')
    # trainer.fit({X: trainX, Y: trainY}, n_epoch=10, show_metric=True, run_id='Summaries_example')

    trainer.fit(feed_dicts={X: trainX, Y: trainY}, n_epoch=1000, val_feed_dicts=0.1,
                shuffle_all=True, show_metric=True, snapshot_step=200,
                snapshot_epoch=False, run_id='alexnet_oxflowers17')

    # Force the weight to be zeros below a threshold
    w_fc1_weights = prune(fc1_weights)
    print("w_fc1_weights =", '\n',  w_fc1_weights, "shape = ", np.shape(w_fc1_weights), '\n')

    w_fc2_weights = prune(fc2_weights)
    print("w_fc2_weights =", '\n',  w_fc2_weights, "shape = ", np.shape(w_fc2_weights), '\n')

    savemat('result_alexnet.mat', {'w_fc1_weights': w_fc1_weights,
                                   'w_fc2_weights': w_fc2_weights,
                                   'learning_rate_': learning_rate_ini,
                                   'lambda_l1': lambda_l1,
                                   'lambda_l2': lambda_l2,
                                   'lambda_dropneuron': lambda_dropneuron,
                                   'dropout': keep_prob})

    # Run the following command to start tensorboard:
    # >> tensorboard /tmp/tflearn_logs/
    # Navigate with your web browser to http://0.0.0.0:6006/