-
Notifications
You must be signed in to change notification settings - Fork 15
/
loggers.py
79 lines (59 loc) · 2.72 KB
/
loggers.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
import os
import torch
from abc import ABC
from torchvision.transforms import ToTensor
to_tensor = ToTensor()
class AbstractLogger(ABC):
def log(self, *args, **kwargs):
raise NotImplementedError
def complete(self, *args, **kwargs):
raise NotImplementedError
def _checkpoint_file_path(export_path, filename):
return os.path.join(export_path, filename)
class RecentModelCheckpointLogger(AbstractLogger):
def __init__(self, export_path, checkpoint_period, ckpt_filename='checkpoint-recent.pth'):
self.export_path = export_path
if not os.path.exists(self.export_path):
os.mkdir(self.export_path)
self.checkpoint_period = checkpoint_period
self.call_count = 0
self.ckpt_filename = ckpt_filename
self.ckpt_final_filename = self.ckpt_filename + '.final'
def log(self, *args, **kwargs):
self.call_count += 1
if self.call_count % self.checkpoint_period == 0:
state_dict = kwargs['state_dict']
state_dict['epoch'] = kwargs['epoch']
state_dict['accum_iter'] = kwargs['accum_iter']
torch.save(state_dict, _checkpoint_file_path(self.export_path, self.ckpt_filename))
def complete(self, *args, **kwargs):
torch.save(kwargs['state_dict'], _checkpoint_file_path(self.export_path, self.ckpt_final_filename))
class BestModelTracker(AbstractLogger):
def __init__(self, export_path, metric_key, ckpt_filename='best_acc_model.pth'):
self.export_path = export_path
if not os.path.exists(self.export_path):
os.mkdir(self.export_path)
self.best_accuracy = 0.
self.metric_key = metric_key
self.ckpt_filename = ckpt_filename
def log(self, *args, **kwargs):
acc = kwargs[self.metric_key]
if self.best_accuracy < acc:
print("Update Best Accuracy Model at {}".format(kwargs['epoch']))
self.best_accuracy = acc
torch.save(kwargs['state_dict'], _checkpoint_file_path(self.export_path, self.ckpt_filename))
def complete(self, *args, **kwargs):
pass
class MetricGraphPrinter(AbstractLogger):
def __init__(self, writer, key='train_loss', graph_label=None, namespace='metric'):
self.key = key
self.graph_label = graph_label if graph_label else self.key
self.group_name = namespace
self.writer = writer
def log(self, *args, **kwargs):
if self.key in kwargs:
self.writer.add_scalar(self.group_name + '/' + self.graph_label, kwargs[self.key], kwargs['accum_iter'])
else:
self.writer.add_scalar(self.group_name + '/' + self.graph_label, 0, kwargs['accum_iter'])
def complete(self, *args, **kwargs):
self.writer.close()