forked from chenyuntc/PyTorchText
-
Notifications
You must be signed in to change notification settings - Fork 0
/
config.py
executable file
·136 lines (114 loc) · 5.46 KB
/
config.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
#coding:utf8
import time
import warnings
tfmt = '%m%d_%H%M%S'
class Config(object):
'''
并不是所有的配置都生效,实际运行中只根据需求获取自己需要的参数
'''
loss = 'multilabelloss'
model='CNNText'
title_dim = 100 # 标题的卷积核数
content_dim = 200 #描述的卷积核数
num_classes = 1999 # 类别
embedding_dim = 256 # embedding大小
linear_hidden_size = 2000 # 全连接层隐藏元数目
kmax_pooling = 2# k
hidden_size = 256 #LSTM hidden size
num_layers=2 #LSTM layers
inception_dim = 512 #inception的卷积核数
# vocab_size = 11973 # num of chars
vocab_size = 411720 # num of words
kernel_size = 3 #单尺度卷积核
kernel_sizes = [2,3,4] #多尺度卷积核
title_seq_len = 50 # 标题长度,word为30 char为50
content_seq_len = 250 #描述长度 word为120 char为250
type_='word' #word 和char
all=False # 模型同时训练char和word
embedding_path = '/mnt/7/zhihu/ieee_zhihu_cup/data/char_embedding.npz' # Embedding
train_data_path = '/mnt/7/zhihu/ieee_zhihu_cup/data/train.npz' # train
labels_path = '/mnt/7/zhihu/ieee_zhihu_cup/data/labels.json' # labels
test_data_path='/mnt/7/zhihu/ieee_zhihu_cup/data/test.npz' # test
result_path='csv/'+time.strftime(tfmt)+'.csv'
shuffle = True # 是否需要打乱数据
num_workers = 4 # 多线程加载所需要的线程数目
pin_memory = True # 数据从CPU->pin_memory—>GPU加速
batch_size = 128
env = time.strftime(tfmt) # Visdom env
plot_every = 10 # 每10个batch，更新visdom等
max_epoch=100
lr = 5e-3 # 学习率
lr2 = 1e-3 # embedding层的学习率
min_lr = 1e-5 # 当学习率低于这个值，就退出训练
lr_decay = 0.99 # 当一个epoch的损失开始上升lr = lr*lr_decay
weight_decay = 0 #2e-5 # 权重衰减
weight = 1 # 正负样本的weight
decay_every = 3000 #每多少个batch 查看一下score,并随之修改学习率
model_path = None # 如果有 就加载
optimizer_path='optimizer.pth' # 优化器的保存地址
debug_file = '/tmp/debug2' #若该文件存在则进入debug模式
debug=False
gpu1 = False #如果在GPU1上运行代码,则需要修改数据存放的路径
floyd=False # 服务如果在floyd上运行需要修改文件路径
zhuge=False # 服务如果在zhuge上运行,修改文件路径
### multimode 用到的
model_names=['MultiCNNTextBNDeep','CNNText_inception','RCNN','LSTMText','CNNText_inception']
model_paths = ['checkpoints/MultiCNNTextBNDeep_0.37125473788','checkpoints/CNNText_tmp_0.380390420742','checkpoints/RCNN_word_0.373609030286','checkpoints/LSTMText_word_0.381833388089','checkpoints/CNNText_tmp_0.376364647145']#,'checkpoints/CNNText_tmp_0.402429167301']
static=False # 是否训练embedding
val=False # 跑测试集还是验证集?
fold = 1 # 数据集fold, 0或1 见 data/fold_dataset.py
augument=True # 是否进行数据增强
###stack
model_num=7
data_root="/data/text/zhihu/result/"
labels_file="/home/a/code/pytorch/zhihu/ddd/labels.json"
val="/home/a/code/pytorch/zhihu/ddd/val.npz"
def parse(self,kwargs,print_=True):
'''
根据字典kwargs 更新 config参数
'''
for k,v in kwargs.iteritems():
if not hasattr(self,k):
raise Exception("opt has not attribute <%s>" %k)
setattr(self,k,v)
###### 根据程序在哪台服务器运行,自动修正数据存放路径 ######
if self.gpu1:
self.train_data_path='/mnt/zhihu/data/train.npz'
self.test_data_path='/mnt/zhihu/data/%s.npz' %('val' if self.val else 'test')
self.labels_path='/mnt/zhihu/data/labels.json'
self.embedding_path=self.embedding_path.replace('/mnt/7/zhihu/ieee_zhihu_cup/','/mnt/zhihu/')
if self.floyd:
self.train_data_path='/data/train.npz'
self.test_data_path='/data/%s.npz' %('val' if self.val else 'test')
self.labels_path='/data/labels.json'
self.embedding_path='/data/char_embedding.npz'
if self.zhuge:
self.train_data_path='./ddd/train.npz'
self.test_data_path='./ddd/%s.npz' %('val' if self.val else 'test')
self.labels_path='./ddd/labels.json'
self.embedding_path='./ddd/char_embedding.npz'
### word和char的长度不一样 ##
if self.type_=='word':
self.vocab_size = 411720 # num of words
self.title_seq_len = 30
self.content_seq_len = 120
self.embedding_path=self.embedding_path.replace('char','word') if self.embedding_path is not None else None
if self.type_=='char':
self.vocab_size = 11973 # num of words
self.title_seq_len = 50
self.content_seq_len = 250
if self.model_path:
self.embedding_path=None
if print_:
print('user config:')
print('#################################')
for k in dir(self):
if not k.startswith('_') and k!='parse' and k!='state_dict':
print k,getattr(self,k)
print('#################################')
return self
def state_dict(self):
return {k:getattr(self,k) for k in dir(self) if not k.startswith('_') and k!='parse' and k!='state_dict' }
Config.parse = parse
Config.state_dict = state_dict
opt = Config()