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
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
|
# Copyright (c) 2018 Samsung Electronics Co., Ltd. All Rights Reserved
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import sys
import traceback
import tensorflow as tf
import model_freezer_util as util
class Tensor(object):
def __init__(self, shape, dtype=tf.float32, const_val=None):
'''
Keyword arguments:
shape -- shaps in a form of list, e.g., 3x2x4 tensor's shape => [3,2,4]. shape of a scalar is []
dtype -- type of values in this Tensor
const_val -- if this Tensor is const, provide this
'''
self._shape = shape
self._const_val = const_val
self._dtype = dtype
def getShape(self):
return self._shape
def getConstVal(self):
return self._const_val
def getDType(self):
return self._dtype
class BaseFreezer(object):
def __init__(self, path):
# files generated by child class will be stored under this path.
self.root_output_path = path
def getTestCases(self):
''' abstract class
override method should return dic containing test cases '''
raise NotImplementedError("please implement this")
def getOutputDirectory(self):
''' abstract class
override method should return directory under self.root_output_path where all pb, pbtxt, checkpoing, tensorboard log are saved '''
raise NotImplementedError("please implement this")
def buildModel(self, sess, test_case_tensor, tc_name):
''' abstract class
override this method to define models when getTestCases() is defined.
this method returns ( input_node_list, output_node_list ) after building a model with all variables set '''
raise NotImplementedError("please implement this")
def createTFInput(self, tensor, input_list):
'''
create TensorFlow tensor and if it is a placeholder, put it into input_list
keyword argument:
tensor -- base_freezer.Tensor
input_list -- vector
'''
if (util.isScalar(tensor)): # check if scalar
if (tensor.getConstVal() == None):
tf_tensor = tf.placeholder(shape=[], dtype=tensor.getDType())
input_list.append(tf_tensor)
else:
tf_tensor = tf.constant(
value=tensor.getConstVal(), dtype=tensor.getDType())
else:
if (tensor.getConstVal() == None):
tf_tensor = tf.placeholder(
shape=tensor.getShape(), dtype=tensor.getDType())
input_list.append(tf_tensor)
else:
tf_tensor = tf.constant(
shape=tensor.getShape(),
value=tensor.getConstVal(),
dtype=tensor.getDType())
return tf_tensor
def saveRelatedFiles(self, sess, input_node_list, output_node_list, fn_prefix):
''' saves pb, pbtxt, chpt files and then freeze graph under top_node_name into directory '''
''' produce pb, pbtxt, and ckpt files '''
(pb_path, pbtxt_path, checkpoint_path) = util.savePbAndCkpt(
sess, self.getOutputDirectory(), fn_prefix)
print("")
print("# 1. Created Tensorflow model files :\n\t-{}\n\t-{}\n\t-{}\n".format(
pb_path, pbtxt_path, checkpoint_path))
'''
produce frozen files
include only nodes below softmax node. nodes for gradient descent (reduce_mean, GradientDescentOptimizer, ...) will not be included
'''
sess.close()
output_node_name = fn_prefix
(frozen_pb_path, frozen_pbtxt_path) = util.freezeGraph(pb_path, checkpoint_path,
output_node_name)
print("")
print("# 2. Freeze() Finished. Created :")
print("\t-{}\n\t-{}\n".format(frozen_pb_path, frozen_pbtxt_path))
self.generateTensorboardLog(pb_path, frozen_pb_path, fn_prefix)
print("")
''' generate tflite file. '''
# manually put back imported modules. refer to https://github.com/tensorflow/tensorflow/issues/15410#issuecomment-352189481
import tempfile
import subprocess
tf.contrib.lite.tempfile = tempfile
tf.contrib.lite.subprocess = subprocess
tflite_path = os.path.join(self.getOutputDirectory(), fn_prefix + ".tflite")
tf.reset_default_graph()
sess = tf.Session()
util.importGraphIntoSession(sess, frozen_pb_path, "")
try:
converter = tf.contrib.lite.TFLiteConverter.from_session(
sess, input_node_list, output_node_list)
tflite_model = converter.convert()
open(tflite_path, "wb").write(tflite_model)
print("# 3. TOCO : Created TFLITE file :\n\t-{}\n".format(tflite_path))
except Exception:
print("# 3. TOCO failed\n")
print(traceback.format_exc())
return (pb_path, frozen_pb_path, tflite_path)
def generateTensorboardLog(self, pb_path, frozen_pb_path, fn_prefix):
''' generating tensorboard logs to compare original pb and frozen pb '''
tensorboardLogDir = util.generateTensorboardLog(
[pb_path, frozen_pb_path], ['original', 'frozen'],
os.path.join(self.getOutputDirectory(), ".tensorboard", fn_prefix))
print("")
print(
"\t# Tensorboard: You can view original graph and frozen graph with tensorboard."
)
print("\t Run the following:")
print("\t $ tensorboard --logdir={} ".format(tensorboardLogDir))
def createSaveFreezeModel(self):
''' method that actually called by main() function. '''
test_cases = self.getTestCases()
# when there are defined test cases
if test_cases != None:
for tc_name in test_cases:
# without this, graph used previous session is reused : https://stackoverflow.com/questions/42706761/closing-session-in-tensorflow-doesnt-reset-graph
tf.reset_default_graph()
# TODO-nnfw session life cycle here is too tangled
sess = tf.Session()
print("")
print("------------ Generating files for {} ------------".format(tc_name))
print("# files will be saved into " + self.getOutputDirectory())
# build model
(input_node_list, output_node_list) = self.buildModel(
sess, test_cases.get(tc_name), tc_name)
''' Now, save to proto buffer format and checkpoint '''
(pb_path, frozen_pb_path, tflite_path) = self.saveRelatedFiles(
sess, input_node_list, output_node_list, tc_name)
sess.close()
# when there is not test cases but the model itself
else:
# without this, graph used previous session is reused : https://stackoverflow.com/questions/42706761/closing-session-in-tensorflow-doesnt-reset-graph
tf.reset_default_graph()
# TODO-nnfw session life cycle here is too tangled
sess = tf.Session()
print("")
print("------------ Generating files for {} ------------".format(tc_name))
print("# files will be saved into " + self.getOutputDirectory())
# build model
(input_node_list, output_node_list) = self.buildModel(
sess, test_cases.get(tc_name), tc_name)
''' Now, save to proto buffer format and checkpoint '''
(pb_path, frozen_pb_path, tflite_path) = self.saveRelatedFiles(
sess, input_node_list, output_node_list, tc_name)
sess.close()
|
