-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfeeder_async.py
209 lines (181 loc) · 8.56 KB
/
feeder_async.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
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
'''
Copyright (C) 2020, Northwestern University and Lawrence Berkeley National Laboratory
See COPYRIGHT notice in top-level directory.
'''
import os
import time
import tensorflow as tf
import yaml
import numpy as np
import h5py
import math
from mpi4py import MPI
import multiprocessing as mp
class cosmoflow_async:
def __init__ (self, yaml_file, lock, cv,
data, label, num_samples,
do_shuffle = 0,
batch_size = 4,
buffer_size = 128):
self.comm = MPI.COMM_WORLD
self.size = self.comm.Get_size()
self.rank = self.comm.Get_rank()
self.lock = lock
self.cv = cv
self.data = data
self.label = label
self.num_samples = num_samples
self.num_buffers = len(data)
self.batch_size = batch_size
self.buffer_size = buffer_size
self.read_index = 0
self.rng = np.random.default_rng()
self.do_shuffle = do_shuffle
self.num_cached_train_batches = 0
self.num_cached_valid_batches = 0
self.train_file_index = 0
self.valid_file_index = 0
self.data_shape = (self.buffer_size, 128, 128, 128, 12)
self.label_shape = (self.buffer_size, 4)
self.file_index = 0
# Parse the given yaml file and get the top dir and file names.
with open (yaml_file, "r") as f:
data = yaml.load(f, Loader = yaml.FullLoader)
for key, value in data.items():
if key == 'frameCnt':
self.samples_per_file = value
self.batches_per_file = int(value / self.batch_size)
if key == 'numPar':
self.label_size = value
if key == 'sourceDir':
self.prj = value['prj']
if key == 'subDir':
self.subdir = value
if key == 'splitIdx':
self.train_files = list(value['train'])
self.valid_files = list(value['val'])
self.train_files = [str(self.prj) + "/" +
str(self.subdir) + "/" +
"PeterA_2019_05_4parE-rec" +
str(file_name[1]) +
".h5" for file_name in enumerate(self.train_files)]
self.valid_files = [str(self.prj) + "/" +
str(self.subdir) + "/" +
"PeterA_2019_05_4parE-rec" +
str(file_name[1]) +
".h5" for file_name in enumerate(self.valid_files)]
print ("Number of samples per file: " + str(self.samples_per_file))
print ("Label size: " + str(self.label_size))
print ("sourceDir.prj: " + str(self.prj))
print ("subDir: " + str(self.subdir))
print ("Buffer size: " + str(self.buffer_size) + " samples")
self.num_train_files = len(self.train_files)
self.offset = int(self.num_train_files / self.size) * self.rank
self.shared_shuffled_index = mp.RawArray('i', self.num_train_files)
# Calculate the number of local training files.
common = int(self.num_train_files / self.size)
remainder = self.num_train_files % self.size
if self.rank < remainder:
self.num_local_train_files = common + 1
else:
self.num_local_train_files = common
# Calculate the number of local validation files.
num_local_valid_files = int(math.floor(len(self.valid_files) / self.size))
local_valid_files_off = num_local_valid_files * self.rank
if self.rank < (len(self.valid_files) % self.size):
num_local_valid_files += 1
local_valid_files_off += self.rank
else:
local_valid_files_off += (len(self.valid_files) % self.size)
self.local_valid_files = self.valid_files[local_valid_files_off:
local_valid_files_off + num_local_valid_files]
# Calculate the number of batches for training and validation.
self.num_train_batches = int(self.batches_per_file * self.num_local_train_files)
self.num_valid_batches = 0
for file_path in self.local_valid_files:
f = h5py.File(file_path, 'r')
self.num_valid_batches += f['unitPar'].shape[0]
f.close()
self.num_valid_batches = int(math.floor(self.num_valid_batches / self.batch_size))
self.shuffle()
def shuffle (self):
# Shuffle the file index.
self.shuffled_file_index = np.arange(self.num_train_files)
self.rng.shuffle(self.shuffled_file_index)
self.comm.Bcast(self.shuffled_file_index, root = 0)
self.shared_shuffled_index[:] = self.shuffled_file_index[:]
self.shuffled_sample_index = np.arange(self.buffer_size)
self.rng.shuffle(self.shuffled_sample_index)
'''
Training dataset
'''
def read_train_sample (self, sample_id):
# Read a random sample from the buffer.
sample_index = sample_id.numpy() % self.buffer_size
sample_index = self.shuffled_sample_index[sample_index]
data_np = np.frombuffer(self.data[self.read_index], dtype = np.uint16).reshape(self.data_shape)
label_np = np.frombuffer(self.label[self.read_index], dtype = np.float32).reshape(self.label_shape)
image = data_np[sample_index]
label = label_np[sample_index]
return image, label
def tf_read_train_sample (self, sample_id):
image, label = tf.py_function(self.read_train_sample, inp=[sample_id], Tout=[tf.float32, tf.float32])
return image, label
def train_dataset (self):
dataset = tf.data.Dataset.from_tensor_slices(np.arange(self.num_local_train_files * self.samples_per_file))
dataset = dataset.map(self.tf_read_train_sample)
dataset = dataset.batch(self.batch_size)
dataset = dataset.repeat()
#dataset = dataset.prefetch(4)
#dataset = dataset.prefetch(tf.data.experimental.AUTOTUNE)
return dataset.__iter__()
def pre_batch (self):
# Wait if the current buffer is empty.
self.lock.acquire()
while self.num_samples[self.read_index].value == 0:
print ("R" + str(self.rank) + " okay, buffer " + str(self.read_index) + " is empty.. I will wait...")
self.cv.notify()
self.cv.wait()
self.lock.release()
if self.num_cached_train_batches == 0:
self.num_cached_train_batches = int(self.buffer_size / self.batch_size)
def post_batch (self):
self.num_cached_train_batches -= 1
if self.num_cached_train_batches == 0:
self.lock.acquire()
self.num_samples[self.read_index].value -= self.buffer_size
self.cv.notify()
self.lock.release()
self.read_index += 1
if self.read_index == self.num_buffers:
self.read_index = 0
'''
Validation dataset
'''
def read_valid_samples (self, batch_id):
# Read a new file if there are no cached batches.
if self.num_cached_valid_batches == 0:
if self.valid_file_index == len(self.local_valid_files):
print ("batch_id: " + str(batch_id) + " Invalid valid_file_index! " + str(self.valid_file_index) + "/" + str(len(self.valid_files)))
f = h5py.File(self.local_valid_files[self.valid_file_index], 'r')
self.valid_file_index += 1
self.images = f['3Dmap'][:]
self.labels = f['unitPar'][:]
f.close()
self.num_cached_valid_batches = int(self.images.shape[0] / self.batch_size)
# Get a mini-batch from the memory buffer.
index = (self.num_cached_valid_batches - 1) * self.batch_size
images = self.images[index : index + self.batch_size]
labels = self.labels[index : index + self.batch_size]
self.num_cached_valid_batches -= 1
return images, labels
def tf_read_valid_samples (self, batch_id):
images, labels = tf.py_function(self.read_valid_samples, inp=[batch_id], Tout=[tf.float32, tf.float32])
images.set_shape([self.batch_size, 128,128,128,12])
labels.set_shape([self.batch_size, 4])
return images, labels
def valid_dataset (self):
dataset = tf.data.Dataset.from_tensor_slices(np.arange(self.num_valid_batches))
dataset = dataset.map(self.tf_read_valid_samples)
dataset = dataset.repeat()
return dataset.__iter__()