-
Notifications
You must be signed in to change notification settings - Fork 0
/
MetricLogger.py
183 lines (162 loc) · 5.74 KB
/
MetricLogger.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
import torch
import torch.distributed as dist
import time
import datetime
from collections import defaultdict, deque
from Distributed import is_dist_avail_and_initialized
class SmoothedValue(object):
"""
keep a series of values and provide access to
- smoothed values over a window
- global average
implement __str__ to expose access to calculated properties
usage
init(fmt=) before training and be added to a metric logger
fmt can contain variables including
- median
- max
- avg(window avg)
- global_avg
- value
more supported variable can be add by adding the corresponding
calculating funcrion and modify the __str__ implementation
"""
def __init__(self, window_size=20, fmt=None):
if fmt is None:
fmt = "{median:.4f} ({global_avg:.4f})"
self.deque = deque(maxlen=window_size)
self.total = 0.0
self.count = 0
self.fmt = fmt
def update(self, batch_mean, batch_size=1):
self.deque.append(batch_mean)
self.count += batch_size
self.total += batch_mean * batch_size
def synchronize_between_process(self):
"""
warning: only synchornize base val count and total
"""
if not is_dist_avail_and_initialized():
return
t = torch.tensor([self.count, self.total], dtype=torch.float64, device='cuda')
dist.barrier()
dist.all_reduce(t)
t = t.tolist()
self.count = int(t[0])
self.total = t[1]
@property
def median(self):
d = torch.tensor(list(self.deque))
return d.median().item()
@property
def avg(self):
d = torch.tensor(list(self.deque), dtype=torch.float32)
return d.mean().item()
@property
def global_avg(self):
return self.total / self.count
@property
def max(self):
return max(self.deque)
@property
def value(self):
return self.deque[-1]
def __str__(self):
return self.fmt.format(
median=self.median,
avg = self.avg,
global_avg=self.global_avg,
max=self.max,
value=self.value)
class MetricLogger(object):
"""
composed of multiple named meters
and support log through their __str__ implementation
main usage:
- add_meter() before training
- update() to save within training
- log_every() to wrap the dataloader loop
"""
def __init__(self, delimiter="\t"):
self.meters = defaultdict(SmoothedValue)
self.delimiter = delimiter
def __getattr__(self, attr):
# enable dict style access to meters
# self.meter_name == self.meters[meter_name]
if attr in self.meters:
return self.meters[attr]
if attr in self.__dict__:
return self.__dict__[attr]
raise AttributeError("'{}' object has no attribute '{}'".format(
type(self).__name__, attr))
def synchronize_between_processes(self):
for meter in self.meters.values():
meter.synchronize_between_processes()
def add_meter(self, name, meter):
self.meters[name] = meter
def update(self, **kwargs):
for k, v in kwargs.items():
if v is None:
continue
if isinstance(v.torch.Tensor):
v = v.item()
assert isinstance(v, (float,int))
self.meters[k].update(v)
def __str__(self):
loss_str = []
for name, meter in self.meters.items():
loss_str.append(
"{}: {}".format(name, str(meter))
)
return self.delimiter.join(loss_str)
def log_every(self, iterable, print_freq, header=None):
"""
serve as an iterator wrapper for dataloader or other iterables
automatically cal
- data_time: at the begin of iteration
- iter_time at the begin of next iteration
"""
i = 0
if not header:
header = ''
start_time = time.time()
end = time.time()
iter_time = SmoothedValue(fmt='{avg:.4f}')
data_time = SmoothedValue(fmt='{avg:.4f}')
space_fmt = ':' + str(len(str(len(iterable)))) + 'd'
log_msg = [
header,
'[{0' + space_fmt + '}/{1}]',
'eta: {eta}',
'{meters}',
'time: {time}',
'data: {data}'
]
if torch.cuda.is_available():
log_msg.append('max mem: {memory:.0f}')
log_msg = self.delimiter.join(log_msg)
MB = 1024.0 * 1024.0
for obj in iterable:
data_time.update(time.time() - end)
yield obj
iter_time.update(time.time() - end)
if i % print_freq == 0 or i == len(iterable) - 1:
eta_seconds = iter_time.global_avg * (len(iterable) - i)
eta_string = str(datetime.timedelta(seconds=int(eta_seconds)))
if torch.cuda.is_available():
print(log_msg.format(
i, len(iterable), eta=eta_string,
meters=str(self),
time=str(iter_time), data=str(data_time),
memory=torch.cuda.max_memory_allocated() / MB))
else:
print(log_msg.format(
i, len(iterable), eta=eta_string,
meters=str(self),
time=str(iter_time), data=str(data_time)))
i += 1
end = time.time()
total_time = time.time() - start_time
total_time_str = str(datetime.timedelta(seconds=int(total_time)))
print('{} Total time: {} ({:.4f} s / it)'.format(
header, total_time_str, total_time / len(iterable)))