-
Notifications
You must be signed in to change notification settings - Fork 14
/
utils.py
191 lines (169 loc) · 6.85 KB
/
utils.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
from keras.layers import *
import jieba
import multiprocessing
import pandas as pd
from gensim.models import Word2Vec
import numpy as np
import keras.backend as K
from keras.callbacks import Callback, ModelCheckpoint
from keras.models import Model
from keras.utils.np_utils import to_categorical
from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
from sklearn.model_selection import StratifiedKFold
from sklearn.metrics import *
import ipykernel
import tensorflow as tf
from sklearn.metrics import roc_auc_score
from sklearn.utils import shuffle
from keras.layers import *
import jieba
import multiprocessing
from gensim.models import Word2Vec
from keras.callbacks import Callback, ModelCheckpoint
from keras.models import Model
from keras.utils.np_utils import to_categorical
from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
from sklearn.model_selection import StratifiedKFold
from sklearn.metrics import *
import ipykernel
from sklearn.metrics import roc_auc_score
from sklearn.utils import shuffle
from tensorflow import set_random_seed
# 设置随机种子
SEED = 2019
np.random.seed(SEED)
set_random_seed(SEED)
def to_text(row, columns):
"""
将类别特征专为文本序列,以空格为间隔
x1 x2 x3
1 2 3
:param row: pandas.DataFrame.row
:param columns: 列名
:return: "x1_1 x2_2 x3_3"
"""
text = []
for col in columns:
text.append(col + '_' + str(row[col]))
return " ".join(text)
def train_w2v(text_list=None, output_vector='data/w2v.txt', embedding_dim=100):
"""
训练word2vec
:param text_list:文本列表
:param output_vector:词向量输出路径
:return:
"""
print("正在训练词向量。。。")
corpus = [text.split() for text in text_list]
model = Word2Vec(corpus, size=embedding_dim, window=5,
iter=20, min_count=1,
workers=multiprocessing.cpu_count()
)
# 保存词向量
model.wv.save_word2vec_format(output_vector, binary=False)
def create_embedding(word_index, w2v_file, embedding_dim, input_length):
"""
# 创建embedding_layer
:param input_length: 序列最大长度
:param embedding_dim: 词向量维度
:param word_index: 词语索引字典
:param w2v_file: 词向量文件
:return:
"""
embedding_index = {}
f = open(w2v_file, 'r', encoding='utf-8')
next(f) # 下一行
for line in f:
values = line.split()
word = values[0]
coefs = np.asarray(values[1:], dtype='float32')
embedding_index[word] = coefs
f.close()
print("Total %d word vectors in w2v_file" % len(embedding_index))
embedding_matrix = np.random.random(size=(len(word_index) + 1, embedding_dim))
for word, i in word_index.items():
embedding_vector = embedding_index.get(word)
if embedding_vector is not None:
embedding_matrix[i] = embedding_vector
embedding_layer = Embedding(len(word_index) + 1,
embedding_dim,
input_length=input_length,
trainable=False)
return embedding_layer
class roc_auc_callback(Callback):
def __init__(self, training_data, validation_data):
super().__init__()
self.x = training_data[0]
self.y = training_data[1]
self.x_val = validation_data[0]
self.y_val = validation_data[1]
def on_train_begin(self, logs=None):
return
def on_train_end(self, logs=None):
return
def on_epoch_begin(self, epoch, logs=None):
return
def on_epoch_end(self, epoch, logs=None):
y_pred = self.model.predict(self.x, verbose=0)
roc = roc_auc_score(self.y, y_pred)
logs['roc_auc'] = roc_auc_score(self.y, y_pred)
logs['norm_gini'] = (roc_auc_score(self.y, y_pred) * 2) - 1
y_pred_val = self.model.predict(self.x_val, verbose=0)
roc_val = roc_auc_score(self.y_val, y_pred_val)
logs['roc_auc_val'] = roc_auc_score(self.y_val, y_pred_val)
logs['norm_gini_val'] = (roc_auc_score(self.y_val, y_pred_val) * 2) - 1
print('\rroc_auc: %s - roc_auc_val: %s - norm_gini: %s - norm_gini_val: %s' % (
str(round(roc, 5)), str(round(roc_val, 5)), str(round((roc * 2 - 1), 5)), str(round((roc_val * 2 - 1), 5))),
end=10 * ' ' + '\n')
return
def on_batch_begin(self, batch, logs=None):
return
def on_batch_end(self, batch, logs=None):
return
def add_noise(series, noise_level):
return series * (1 + noise_level * np.random.randn(len(series)))
def target_encode(trn_series=None,
tst_series=None,
target=None,
min_samples_leaf=1,
smoothing=1,
noise_level=0):
"""
Smoothing is computed like in the following paper by Daniele Micci-Barreca
https://kaggle2.blob.core.windows.net/forum-message-attachments/225952/7441/high%20cardinality%20categoricals.pdf
trn_series : training categorical feature as a pd.Series
tst_series : test categorical feature as a pd.Series
target : target data as a pd.Series
min_samples_leaf (int) : minimum samples to take category average into account
smoothing (int) : smoothing effect to balance categorical average vs prior
"""
assert len(trn_series) == len(target)
assert trn_series.name == tst_series.name
temp = pd.concat([trn_series, target], axis=1)
# Compute target mean
averages = temp.groupby(by=trn_series.name)[target.name].agg(["mean", "count"])
# Compute smoothing
smoothing = 1 / (1 + np.exp(-(averages["count"] - min_samples_leaf) / smoothing))
# Apply average function to all target data
prior = target.mean()
# The bigger the count the less full_avg is taken into account
averages[target.name] = prior * (1 - smoothing) + averages["mean"] * smoothing
averages.drop(["mean", "count"], axis=1, inplace=True)
# Apply averages to trn and tst series
ft_trn_series = pd.merge(
trn_series.to_frame(trn_series.name),
averages.reset_index().rename(columns={'index': target.name, target.name: 'average'}),
on=trn_series.name,
how='left')['average'].rename(trn_series.name + '_mean').fillna(prior)
# pd.merge does not keep the index so restore it
ft_trn_series.index = trn_series.index
ft_tst_series = pd.merge(
tst_series.to_frame(tst_series.name),
averages.reset_index().rename(columns={'index': target.name, target.name: 'average'}),
on=tst_series.name,
how='left')['average'].rename(trn_series.name + '_mean').fillna(prior)
# pd.merge does not keep the index so restore it
ft_tst_series.index = tst_series.index
return add_noise(ft_trn_series, noise_level), add_noise(ft_tst_series, noise_level)