res_DeepCrossing模型
DeepCrossing
4. 代码实现
# 导包
import warnings
warnings.filterwarnings("ignore")
import pandas as pd
import numpy as np
import tensorflow as tf
from tensorflow.keras.layers import *
from tensorflow.keras.models import *
from sklearn.preprocessing import MinMaxScaler, LabelEncoder
# from utils import SparseFeat, DenseFeat, VarLenSparseFeat
# from collections import namedtuple
# # 使用具名元组定义特征标记
# SparseFeat = namedtuple('SparseFeat', ['name', 'vocabulary_size', 'embedding_dim'])
# DenseFeat = namedtuple('DenseFeat', ['name', 'dimension'])
# VarLenSparseFeat = namedtuple('VarLenSparseFeat', ['name', 'vocabulary_size', 'embedding_dim', 'maxlen'])
1.读取并处理数据
- 读取数据
- 划分出数值型特征和类别型特征
- 数值型特征处理 – 控制填充,取对数(归一化?)
- 类别型特征的处理 – one-hot编码
# 读取数据
data = pd.read_csv('./data/criteo_sample.txt')
data.head()
| label | I1 | I2 | I3 | I4 | I5 | I6 | I7 | I8 | I9 | ... | C17 | C18 | C19 | C20 | C21 | C22 | C23 | C24 | C25 | C26 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | NaN | 3 | 260.0 | NaN | 17668.0 | NaN | NaN | 33.0 | NaN | ... | e5ba7672 | 87c6f83c | NaN | NaN | 0429f84b | NaN | 3a171ecb | c0d61a5c | NaN | NaN |
| 1 | 0 | NaN | -1 | 19.0 | 35.0 | 30251.0 | 247.0 | 1.0 | 35.0 | 160.0 | ... | d4bb7bd8 | 6fc84bfb | NaN | NaN | 5155d8a3 | NaN | be7c41b4 | ded4aac9 | NaN | NaN |
| 2 | 0 | 0.0 | 0 | 2.0 | 12.0 | 2013.0 | 164.0 | 6.0 | 35.0 | 523.0 | ... | e5ba7672 | 675c9258 | NaN | NaN | 2e01979f | NaN | bcdee96c | 6d5d1302 | NaN | NaN |
| 3 | 0 | NaN | 13 | 1.0 | 4.0 | 16836.0 | 200.0 | 5.0 | 4.0 | 29.0 | ... | e5ba7672 | 52e44668 | NaN | NaN | e587c466 | NaN | 32c7478e | 3b183c5c | NaN | NaN |
| 4 | 0 | 0.0 | 0 | 104.0 | 27.0 | 1990.0 | 142.0 | 4.0 | 32.0 | 37.0 | ... | e5ba7672 | 25c88e42 | 21ddcdc9 | b1252a9d | 0e8585d2 | NaN | 32c7478e | 0d4a6d1a | 001f3601 | 92c878de |
5 rows × 40 columns
data.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 200 entries, 0 to 199
Data columns (total 40 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 label 200 non-null int64
1 I1 110 non-null float64
2 I2 200 non-null int64
3 I3 166 non-null float64
4 I4 165 non-null float64
5 I5 194 non-null float64
6 I6 149 non-null float64
7 I7 190 non-null float64
8 I8 200 non-null float64
9 I9 190 non-null float64
10 I10 110 non-null float64
11 I11 190 non-null float64
12 I12 43 non-null float64
13 I13 165 non-null float64
14 C1 200 non-null object
15 C2 200 non-null object
16 C3 191 non-null object
17 C4 191 non-null object
18 C5 200 non-null object
19 C6 168 non-null object
20 C7 200 non-null object
21 C8 200 non-null object
22 C9 200 non-null object
23 C10 200 non-null object
24 C11 200 non-null object
25 C12 191 non-null object
26 C13 200 non-null object
27 C14 200 non-null object
28 C15 200 non-null object
29 C16 191 non-null object
30 C17 200 non-null object
31 C18 200 non-null object
32 C19 118 non-null object
33 C20 118 non-null object
34 C21 191 non-null object
35 C22 41 non-null object
36 C23 200 non-null object
37 C24 191 non-null object
38 C25 118 non-null object
39 C26 118 non-null object
dtypes: float64(12), int64(2), object(26)
memory usage: 62.6+ KB
# 将数据划分为sparse_feature和dense_feature
columns = data.columns.values
columns
array(['label', 'I1', 'I2', 'I3', 'I4', 'I5', 'I6', 'I7', 'I8', 'I9',
'I10', 'I11', 'I12', 'I13', 'C1', 'C2', 'C3', 'C4', 'C5', 'C6',
'C7', 'C8', 'C9', 'C10', 'C11', 'C12', 'C13', 'C14', 'C15', 'C16',
'C17', 'C18', 'C19', 'C20', 'C21', 'C22', 'C23', 'C24', 'C25',
'C26'], dtype=object)
dense_features = [ feat for feat in columns if 'I' in feat]
sparse_features = [feat for feat in columns if 'C' in feat]
dense_features+sparse_features # 列表可以相加
['I1',
'I2',
'I3',
'I4',
'I5',
'I6',
'I7',
'I8',
'I9',
'I10',
'I11',
'I12',
'I13',
'C1',
'C2',
'C3',
'C4',
'C5',
'C6',
'C7',
'C8',
'C9',
'C10',
'C11',
'C12',
'C13',
'C14',
'C15',
'C16',
'C17',
'C18',
'C19',
'C20',
'C21',
'C22',
'C23',
'C24',
'C25',
'C26']
# 数据处理的函数
def data_process(data_df,dense_features,sparse_features):
"""
简单处理特征，包括填充缺失值，数值处理，类别编码
param data_df: DataFrame格式的数据
param dense_features: 数值特征名称列表
param sparse_features: 类别特征名称列表
"""
data_df[dense_features] = data_df[dense_features].fillna(0.0) # 空值补0.0
for f in dense_features: # 对数值型特征的处理
data_df[f] = data_df[f].apply(lambda x: np.log(x+1) if x > -1 else -1)
# 将类别型特征的空值填上-1
data_df[sparse_features] = data_df[sparse_features].fillna("-1")
for f in sparse_features:
lbe = LabelEncoder() # 对类别型特征进行了one-hot
data_df[f] = lbe.fit_transform(data_df[f]) # 传入的是个列表
return data_df[dense_features + sparse_features]
train_data = data_process(data, dense_features, sparse_features)
train_data
| I1 | I2 | I3 | I4 | I5 | I6 | I7 | I8 | I9 | I10 | ... | C17 | C18 | C19 | C20 | C21 | C22 | C23 | C24 | C25 | C26 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.000000 | 1.386294 | 5.564520 | 0.000000 | 9.779567 | 0.000000 | 0.000000 | 3.526361 | 0.000000 | 0.000000 | ... | 8 | 66 | 0 | 0 | 3 | 0 | 1 | 96 | 0 | 0 |
| 1 | 0.000000 | -1.000000 | 2.995732 | 3.583519 | 10.317318 | 5.513429 | 0.693147 | 3.583519 | 5.081404 | 0.000000 | ... | 7 | 52 | 0 | 0 | 47 | 0 | 7 | 112 | 0 | 0 |
| 2 | 0.000000 | 0.000000 | 1.098612 | 2.564949 | 7.607878 | 5.105945 | 1.945910 | 3.583519 | 6.261492 | 0.000000 | ... | 8 | 49 | 0 | 0 | 25 | 0 | 6 | 53 | 0 | 0 |
| 3 | 0.000000 | 2.639057 | 0.693147 | 1.609438 | 9.731334 | 5.303305 | 1.791759 | 1.609438 | 3.401197 | 0.000000 | ... | 8 | 37 | 0 | 0 | 156 | 0 | 0 | 32 | 0 | 0 |
| 4 | 0.000000 | 0.000000 | 4.653960 | 3.332205 | 7.596392 | 4.962845 | 1.609438 | 3.496508 | 3.637586 | 0.000000 | ... | 8 | 14 | 5 | 3 | 9 | 0 | 0 | 5 | 1 | 47 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 195 | 0.000000 | 0.000000 | 4.736198 | 1.386294 | 8.018625 | 6.356108 | 1.098612 | 1.386294 | 5.370638 | 0.000000 | ... | 0 | 74 | 5 | 1 | 30 | 5 | 0 | 118 | 17 | 48 |
| 196 | 0.000000 | 0.693147 | 0.693147 | 0.693147 | 7.382746 | 2.564949 | 0.693147 | 2.564949 | 2.772589 | 0.000000 | ... | 1 | 25 | 0 | 0 | 138 | 0 | 0 | 68 | 0 | 0 |
| 197 | 0.693147 | 0.000000 | 1.945910 | 1.386294 | 0.000000 | 0.000000 | 2.995732 | 1.386294 | 1.386294 | 0.693147 | ... | 4 | 40 | 17 | 2 | 41 | 0 | 0 | 12 | 16 | 11 |
| 198 | 0.000000 | 3.135494 | 1.945910 | 3.135494 | 5.318120 | 5.036953 | 4.394449 | 2.944439 | 6.232448 | 0.000000 | ... | 4 | 7 | 18 | 1 | 123 | 0 | 0 | 10 | 16 | 49 |
| 199 | 0.693147 | -1.000000 | 0.000000 | 0.000000 | 4.934474 | 0.000000 | 0.693147 | 0.000000 | 0.000000 | 0.693147 | ... | 7 | 72 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
200 rows × 39 columns
train_data['label'] = data['label']
train_data
| I1 | I2 | I3 | I4 | I5 | I6 | I7 | I8 | I9 | I10 | ... | C18 | C19 | C20 | C21 | C22 | C23 | C24 | C25 | C26 | label | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.000000 | 1.386294 | 5.564520 | 0.000000 | 9.779567 | 0.000000 | 0.000000 | 3.526361 | 0.000000 | 0.000000 | ... | 66 | 0 | 0 | 3 | 0 | 1 | 96 | 0 | 0 | 0 |
| 1 | 0.000000 | -1.000000 | 2.995732 | 3.583519 | 10.317318 | 5.513429 | 0.693147 | 3.583519 | 5.081404 | 0.000000 | ... | 52 | 0 | 0 | 47 | 0 | 7 | 112 | 0 | 0 | 0 |
| 2 | 0.000000 | 0.000000 | 1.098612 | 2.564949 | 7.607878 | 5.105945 | 1.945910 | 3.583519 | 6.261492 | 0.000000 | ... | 49 | 0 | 0 | 25 | 0 | 6 | 53 | 0 | 0 | 0 |
| 3 | 0.000000 | 2.639057 | 0.693147 | 1.609438 | 9.731334 | 5.303305 | 1.791759 | 1.609438 | 3.401197 | 0.000000 | ... | 37 | 0 | 0 | 156 | 0 | 0 | 32 | 0 | 0 | 0 |
| 4 | 0.000000 | 0.000000 | 4.653960 | 3.332205 | 7.596392 | 4.962845 | 1.609438 | 3.496508 | 3.637586 | 0.000000 | ... | 14 | 5 | 3 | 9 | 0 | 0 | 5 | 1 | 47 | 0 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 195 | 0.000000 | 0.000000 | 4.736198 | 1.386294 | 8.018625 | 6.356108 | 1.098612 | 1.386294 | 5.370638 | 0.000000 | ... | 74 | 5 | 1 | 30 | 5 | 0 | 118 | 17 | 48 | 0 |
| 196 | 0.000000 | 0.693147 | 0.693147 | 0.693147 | 7.382746 | 2.564949 | 0.693147 | 2.564949 | 2.772589 | 0.000000 | ... | 25 | 0 | 0 | 138 | 0 | 0 | 68 | 0 | 0 | 1 |
| 197 | 0.693147 | 0.000000 | 1.945910 | 1.386294 | 0.000000 | 0.000000 | 2.995732 | 1.386294 | 1.386294 | 0.693147 | ... | 40 | 17 | 2 | 41 | 0 | 0 | 12 | 16 | 11 | 1 |
| 198 | 0.000000 | 3.135494 | 1.945910 | 3.135494 | 5.318120 | 5.036953 | 4.394449 | 2.944439 | 6.232448 | 0.000000 | ... | 7 | 18 | 1 | 123 | 0 | 0 | 10 | 16 | 49 | 0 |
| 199 | 0.693147 | -1.000000 | 0.000000 | 0.000000 | 4.934474 | 0.000000 | 0.693147 | 0.000000 | 0.000000 | 0.693147 | ... | 72 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
200 rows × 40 columns
train_data.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 200 entries, 0 to 199
Data columns (total 40 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 I1 200 non-null float64
1 I2 200 non-null float64
2 I3 200 non-null float64
3 I4 200 non-null float64
4 I5 200 non-null float64
5 I6 200 non-null float64
6 I7 200 non-null float64
7 I8 200 non-null float64
8 I9 200 non-null float64
9 I10 200 non-null float64
10 I11 200 non-null float64
11 I12 200 non-null float64
12 I13 200 non-null float64
13 C1 200 non-null int32
14 C2 200 non-null int32
15 C3 200 non-null int32
16 C4 200 non-null int32
17 C5 200 non-null int32
18 C6 200 non-null int32
19 C7 200 non-null int32
20 C8 200 non-null int32
21 C9 200 non-null int32
22 C10 200 non-null int32
23 C11 200 non-null int32
24 C12 200 non-null int32
25 C13 200 non-null int32
26 C14 200 non-null int32
27 C15 200 non-null int32
28 C16 200 non-null int32
29 C17 200 non-null int32
30 C18 200 non-null int32
31 C19 200 non-null int32
32 C20 200 non-null int32
33 C21 200 non-null int32
34 C22 200 non-null int32
35 C23 200 non-null int32
36 C24 200 non-null int32
37 C25 200 non-null int32
38 C26 200 non-null int32
39 label 200 non-null int64
dtypes: float64(13), int32(26), int64(1)
memory usage: 42.3 KB
dense_features
['I1',
'I2',
'I3',
'I4',
'I5',
'I6',
'I7',
'I8',
'I9',
'I10',
'I11',
'I12',
'I13']
2.模型的创建开始了
2.1数值型特征的处理
- 数值型特征的读入
- 拼接数值型特征
dense_inputs = [] # 用来存放数值型特征的输入
for f in dense_features:
_input = Input([1], name=f)
dense_inputs.append(_input)
dense_inputs
[<tf.Tensor 'I1:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I2:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I3:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I4:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I5:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I6:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I7:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I8:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I9:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I10:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I11:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I12:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I13:0' shape=(None, 1) dtype=float32>]
dense_concat_inputs = Concatenate(axis=1)(dense_inputs)
dense_concat_inputs
<tf.Tensor 'concatenate/concat:0' shape=(None, 13) dtype=float32>
2.2 类别型特征
- 类别型特征的输入
- 转化为Embedding
- 拼接类别型特征
sparse_inputs = [] # 用来存放类别型特征的输入
for f in sparse_features:
_input = Input([1], name=f)
sparse_inputs.append(_input)
sparse_inputs
[<tf.Tensor 'C1:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C2:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C3:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C4:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C5:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C6:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C7:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C8:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C9:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C10:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C11:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C12:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C13:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C14:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C15:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C16:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C17:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C18:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C19:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C20:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C21:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C22:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C23:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C24:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C25:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C26:0' shape=(None, 1) dtype=float32>]
- 对类别型特征进行emb
sparse_1d_embed = [] # 存放各个类别型特征对应的Embedding后的结果
for i, _input in enumerate(sparse_inputs):
f = sparse_features[i] # f : C1 ... C26, sparse_features是个列表
voc_size = train_data[f].nunique()
_embed = Flatten()(Embedding(voc_size+1, 4,)(_input))
sparse_1d_embed.append(_embed)
sparse_1d_embed
[<tf.Tensor 'flatten_26/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_27/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_28/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_29/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_30/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_31/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_32/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_33/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_34/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_35/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_36/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_37/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_38/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_39/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_40/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_41/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_42/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_43/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_44/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_45/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_46/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_47/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_48/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_49/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_50/Reshape:0' shape=(None, 4) dtype=float32>,
<tf.Tensor 'flatten_51/Reshape:0' shape=(None, 4) dtype=float32>]
# 类别型特征的拼接
sparse_1d_embed = Concatenate(axis=1)(sparse_1d_embed) # B x m*dim (n表示类别特征的数量，dim表示embedding的维度)
sparse_1d_embed
<tf.Tensor 'concatenate_1/concat:0' shape=(None, 104) dtype=float32>
2.3数值型特征和类别型特征拼接到一起
# 将dense特征和Sparse特征拼接到一起
all_inputs = Concatenate(axis=1)([dense_concat_inputs, sparse_1d_embed]) # B x (n + m*dim)
all_inputs
<tf.Tensor 'concatenate_2/concat:0' shape=(None, 117) dtype=float32>
2.4 残差
# DNN残差块的定义
class ResidualBlock(Layer):
def __init__(self, units): # units表示的是DNN隐藏层神经元数量
super(ResidualBlock, self).__init__()
self.units = units
def build(self, input_shape):
out_dim = input_shape[-1]
self.dnn1 = Dense(self.units, activation='relu')
self.dnn2 = Dense(out_dim, activation='relu') # 保证输入的维度和输出的维度一致才能进行残差连接
def call(self, inputs):
x = inputs
x = self.dnn1(x)
x = self.dnn2(x)
x = Activation('relu')(x + inputs) # 残差操作
return x
# block_nums表示DNN残差块的数量
def get_dnn_logits(dnn_inputs, block_nums=3):
dnn_out = dnn_inputs
for i in range(block_nums):
dnn_out = ResidualBlock(64)(dnn_out)
# 将dnn的输出转化成logits
dnn_logits = Dense(1, activation='sigmoid')(dnn_out)
return dnn_logits
# 输入到dnn中，需要提前定义需要几个残差块
output_layer = get_dnn_logits(all_inputs, block_nums=3)
output_layer
<tf.Tensor 'dense/Sigmoid:0' shape=(None, 1) dtype=float32>
dense_inputs+sparse_inputs
[<tf.Tensor 'I1:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I2:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I3:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I4:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I5:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I6:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I7:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I8:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I9:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I10:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I11:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I12:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'I13:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C1:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C2:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C3:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C4:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C5:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C6:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C7:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C8:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C9:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C10:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C11:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C12:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C13:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C14:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C15:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C16:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C17:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C18:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C19:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C20:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C21:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C22:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C23:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C24:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C25:0' shape=(None, 1) dtype=float32>,
<tf.Tensor 'C26:0' shape=(None, 1) dtype=float32>]
model = Model(dense_inputs+sparse_inputs, output_layer)
model.summary()
Model: "functional_1"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
C1 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C2 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C3 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C4 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C5 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C6 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C7 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C8 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C9 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C10 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C11 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C12 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C13 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C14 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C15 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C16 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C17 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C18 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C19 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C20 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C21 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C22 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C23 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C24 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C25 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
C26 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
embedding (Embedding) (None, 1, 4) 112 C1[0][0]
__________________________________________________________________________________________________
embedding_1 (Embedding) (None, 1, 4) 372 C2[0][0]
__________________________________________________________________________________________________
embedding_2 (Embedding) (None, 1, 4) 692 C3[0][0]
__________________________________________________________________________________________________
embedding_3 (Embedding) (None, 1, 4) 632 C4[0][0]
__________________________________________________________________________________________________
embedding_4 (Embedding) (None, 1, 4) 52 C5[0][0]
__________________________________________________________________________________________________
embedding_5 (Embedding) (None, 1, 4) 32 C6[0][0]
__________________________________________________________________________________________________
embedding_6 (Embedding) (None, 1, 4) 736 C7[0][0]
__________________________________________________________________________________________________
embedding_7 (Embedding) (None, 1, 4) 80 C8[0][0]
__________________________________________________________________________________________________
embedding_8 (Embedding) (None, 1, 4) 12 C9[0][0]
__________________________________________________________________________________________________
embedding_9 (Embedding) (None, 1, 4) 572 C10[0][0]
__________________________________________________________________________________________________
embedding_10 (Embedding) (None, 1, 4) 696 C11[0][0]
__________________________________________________________________________________________________
embedding_11 (Embedding) (None, 1, 4) 684 C12[0][0]
__________________________________________________________________________________________________
embedding_12 (Embedding) (None, 1, 4) 668 C13[0][0]
__________________________________________________________________________________________________
embedding_13 (Embedding) (None, 1, 4) 60 C14[0][0]
__________________________________________________________________________________________________
embedding_14 (Embedding) (None, 1, 4) 684 C15[0][0]
__________________________________________________________________________________________________
embedding_15 (Embedding) (None, 1, 4) 676 C16[0][0]
__________________________________________________________________________________________________
embedding_16 (Embedding) (None, 1, 4) 40 C17[0][0]
__________________________________________________________________________________________________
embedding_17 (Embedding) (None, 1, 4) 512 C18[0][0]
__________________________________________________________________________________________________
embedding_18 (Embedding) (None, 1, 4) 180 C19[0][0]
__________________________________________________________________________________________________
embedding_19 (Embedding) (None, 1, 4) 20 C20[0][0]
__________________________________________________________________________________________________
embedding_20 (Embedding) (None, 1, 4) 680 C21[0][0]
__________________________________________________________________________________________________
embedding_21 (Embedding) (None, 1, 4) 28 C22[0][0]
__________________________________________________________________________________________________
embedding_22 (Embedding) (None, 1, 4) 44 C23[0][0]
__________________________________________________________________________________________________
embedding_23 (Embedding) (None, 1, 4) 504 C24[0][0]
__________________________________________________________________________________________________
embedding_24 (Embedding) (None, 1, 4) 84 C25[0][0]
__________________________________________________________________________________________________
embedding_25 (Embedding) (None, 1, 4) 364 C26[0][0]
__________________________________________________________________________________________________
I1 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I2 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I3 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I4 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I5 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I6 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I7 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I8 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I9 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I10 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I11 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I12 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
I13 (InputLayer) [(None, 1)] 0
__________________________________________________________________________________________________
flatten (Flatten) (None, 4) 0 embedding[0][0]
__________________________________________________________________________________________________
flatten_1 (Flatten) (None, 4) 0 embedding_1[0][0]
__________________________________________________________________________________________________
flatten_2 (Flatten) (None, 4) 0 embedding_2[0][0]
__________________________________________________________________________________________________
flatten_3 (Flatten) (None, 4) 0 embedding_3[0][0]
__________________________________________________________________________________________________
flatten_4 (Flatten) (None, 4) 0 embedding_4[0][0]
__________________________________________________________________________________________________
flatten_5 (Flatten) (None, 4) 0 embedding_5[0][0]
__________________________________________________________________________________________________
flatten_6 (Flatten) (None, 4) 0 embedding_6[0][0]
__________________________________________________________________________________________________
flatten_7 (Flatten) (None, 4) 0 embedding_7[0][0]
__________________________________________________________________________________________________
flatten_8 (Flatten) (None, 4) 0 embedding_8[0][0]
__________________________________________________________________________________________________
flatten_9 (Flatten) (None, 4) 0 embedding_9[0][0]
__________________________________________________________________________________________________
flatten_10 (Flatten) (None, 4) 0 embedding_10[0][0]
__________________________________________________________________________________________________
flatten_11 (Flatten) (None, 4) 0 embedding_11[0][0]
__________________________________________________________________________________________________
flatten_12 (Flatten) (None, 4) 0 embedding_12[0][0]
__________________________________________________________________________________________________
flatten_13 (Flatten) (None, 4) 0 embedding_13[0][0]
__________________________________________________________________________________________________
flatten_14 (Flatten) (None, 4) 0 embedding_14[0][0]
__________________________________________________________________________________________________
flatten_15 (Flatten) (None, 4) 0 embedding_15[0][0]
__________________________________________________________________________________________________
flatten_16 (Flatten) (None, 4) 0 embedding_16[0][0]
__________________________________________________________________________________________________
flatten_17 (Flatten) (None, 4) 0 embedding_17[0][0]
__________________________________________________________________________________________________
flatten_18 (Flatten) (None, 4) 0 embedding_18[0][0]
__________________________________________________________________________________________________
flatten_19 (Flatten) (None, 4) 0 embedding_19[0][0]
__________________________________________________________________________________________________
flatten_20 (Flatten) (None, 4) 0 embedding_20[0][0]
__________________________________________________________________________________________________
flatten_21 (Flatten) (None, 4) 0 embedding_21[0][0]
__________________________________________________________________________________________________
flatten_22 (Flatten) (None, 4) 0 embedding_22[0][0]
__________________________________________________________________________________________________
flatten_23 (Flatten) (None, 4) 0 embedding_23[0][0]
__________________________________________________________________________________________________
flatten_24 (Flatten) (None, 4) 0 embedding_24[0][0]
__________________________________________________________________________________________________
flatten_25 (Flatten) (None, 4) 0 embedding_25[0][0]
__________________________________________________________________________________________________
concatenate (Concatenate) (None, 13) 0 I1[0][0]
I2[0][0]
I3[0][0]
I4[0][0]
I5[0][0]
I6[0][0]
I7[0][0]
I8[0][0]
I9[0][0]
I10[0][0]
I11[0][0]
I12[0][0]
I13[0][0]
__________________________________________________________________________________________________
concatenate_1 (Concatenate) (None, 104) 0 flatten[0][0]
flatten_1[0][0]
flatten_2[0][0]
flatten_3[0][0]
flatten_4[0][0]
flatten_5[0][0]
flatten_6[0][0]
flatten_7[0][0]
flatten_8[0][0]
flatten_9[0][0]
flatten_10[0][0]
flatten_11[0][0]
flatten_12[0][0]
flatten_13[0][0]
flatten_14[0][0]
flatten_15[0][0]
flatten_16[0][0]
flatten_17[0][0]
flatten_18[0][0]
flatten_19[0][0]
flatten_20[0][0]
flatten_21[0][0]
flatten_22[0][0]
flatten_23[0][0]
flatten_24[0][0]
flatten_25[0][0]
__________________________________________________________________________________________________
concatenate_2 (Concatenate) (None, 117) 0 concatenate[0][0]
concatenate_1[0][0]
__________________________________________________________________________________________________
residual_block (ResidualBlock) (None, 117) 15157 concatenate_2[0][0]
__________________________________________________________________________________________________
residual_block_1 (ResidualBlock (None, 117) 15157 residual_block[0][0]
__________________________________________________________________________________________________
residual_block_2 (ResidualBlock (None, 117) 15157 residual_block_1[0][0]
__________________________________________________________________________________________________
dense (Dense) (None, 1) 118 residual_block_2[0][0]
==================================================================================================
Total params: 54,805
Trainable params: 54,805
Non-trainable params: 0
__________________________________________________________________________________________________
model.compile(optimizer="adam",
loss="binary_crossentropy",
metrics=[tf.keras.metrics.AUC(name='auc')])
# 将输入数据转化成字典的形式输入 -- 这样可以
train_model_input = {name: data[name] for name in dense_features + sparse_features}
# 这样还不行,这是个列表
# train_model_input_1 = [data[name] for name in dense_features + sparse_features]
# type(train_model_input_1)
list
# array数组可以
train_dense_x = [train_data[f].values for f in dense_features]
train_sparse_x = [train_data[f].values for f in sparse_features]
train_dense_x
[array([0. , 0. , 0. , 0. , 0. ,
0. , 0. , 2.99573227, 0. , 1.09861229,
1.09861229, 0. , 0. , 2.19722458, 0. ,
0. , 0. , 0. , 1.79175947, 1.09861229,
0. , 0. , 0. , 0. , 0. ,
2.30258509, 0. , 0. , 2.19722458, 0. ,
0.69314718, 0.69314718, 2.48490665, 2.39789527, 0. ,
0. , 0. , 0. , 0. , 2.39789527,
0. , 0. , 0. , 0.69314718, 1.09861229,
0. , 1.60943791, 1.09861229, 0. , 0. ,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 2.07944154, 0. , 1.79175947, 0. ,
0. , 0. , 0. , 1.09861229, 0. ,
1.60943791, 0. , 0.69314718, 0. , 0.69314718,
0. , 0. , 0. , 0. , 0.69314718,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 0. , 0. , 0. , 1.38629436,
0. , 0. , 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0.69314718, 1.38629436,
0. , 0. , 0. , 2.19722458, 0. ,
0. , 0. , 0. , 0. , 1.38629436,
0. , 0. , 3.63758616, 1.79175947, 0. ,
0. , 0. , 0. , 1.38629436, 0. ,
0. , 1.09861229, 0.69314718, 0. , 2.07944154,
0. , 0. , 0. , 0. , 1.09861229,
0. , 2.56494936, 0. , 0. , 0. ,
0. , 0. , 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.69314718, 2.30258509, 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0. , 1.60943791,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 2.19722458, 0. , 0. , 0. ,
0. , 0. , 0. , 0.69314718, 0.69314718,
0.69314718, 0. , 1.09861229, 0.69314718, 0. ,
0. , 1.79175947, 0. , 0. , 0. ,
0. , 1.60943791, 0. , 0. , 1.09861229,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.69314718, 0. , 0.69314718]),
array([ 1.38629436, -1. , 0. , 2.63905733, 0. ,
-1. , 5.91620206, 2.39789527, 0. , 2.48490665,
0.69314718, 1.09861229, 1.79175947, -1. , 0. ,
4.41884061, 3.21887582, 4.66343909, 4.4543473 , 1.38629436,
0.69314718, 1.60943791, 0. , 3.78418963, 0.69314718,
0.69314718, 0.69314718, 3.29583687, -1. , 0.69314718,
5.71373281, 0. , 5.52942909, 0.69314718, -1. ,
3.58351894, 0.69314718, 0. , 5.36129217, 2.48490665,
1.79175947, 6.65929392, 5.19295685, 0.69314718, 4.29045944,
4.06044301, 0.69314718, 1.60943791, 1.79175947, 3.91202301,
7.96067261, 4.78749174, 3.25809654, 5.19849703, 1.09861229,
0. , 4.44265126, 0. , 0.69314718, 2.07944154,
4.00733319, 0. , 0.69314718, 1.79175947, 2.99573227,
0. , 1.79175947, 0.69314718, 0.69314718, -1. ,
2.94443898, 1.79175947, 3.33220451, 3.91202301, 2.99573227,
0.69314718, 0.69314718, 4.07753744, 2.48490665, 7.84815309,
3.4339872 , 3.91202301, -1. , 2.07944154, 4.2341065 ,
5.72031178, 1.09861229, 0. , 0. , 3.13549422,
0.69314718, 0.69314718, 1.60943791, 4.17438727, 0. ,
0.69314718, 3.98898405, 0. , 0.69314718, 0.69314718,
0. , 0. , 6.2126061 , 2.07944154, 1.60943791,
1.94591015, 4.06044301, 4.51085951, 3.40119738, -1. ,
1.09861229, 4.36944785, 4.73619845, 0.69314718, 0. ,
0.69314718, 0.69314718, 1.79175947, 1.09861229, 2.19722458,
0. , 0.69314718, 7.52294092, 4.18965474, 5.10594547,
2.39789527, 2.77258872, -1. , 1.38629436, 0. ,
0.69314718, 0.69314718, 5.03043792, 0.69314718, -1. ,
2.7080502 , -1. , 1.09861229, 2.89037176, 0.69314718,
5.21493576, 2.63905733, 0.69314718, 0.69314718, 0.69314718,
1.09861229, 0.69314718, 2.56494936, 0. , -1. ,
5.91620206, 5.47227067, 0. , 1.38629436, 0.69314718,
2.39789527, 1.09861229, 0.69314718, 0.69314718, 0.69314718,
0.69314718, 0.69314718, -1. , 4.31748811, 3.66356165,
0.69314718, 2.7080502 , 2.94443898, 1.09861229, 0. ,
7.98002359, 4.02535169, 2.19722458, 6.20253552, 3.58351894,
2.19722458, 0. , 0. , 5.5174529 , 0. ,
0. , -1. , 3.66356165, 0. , 1.94591015,
3.33220451, 8.00703401, 5.19295685, 0.69314718, 0. ,
0. , 1.09861229, 2.89037176, 1.09861229, 2.99573227,
0. , 0.69314718, 0. , 3.13549422, -1. ]),
array([5.56452041, 2.99573227, 1.09861229, 0.69314718, 4.65396035,
4.15888308, 1.60943791, 3.4339872 , 3.61091791, 2.19722458,
5.25227343, 1.09861229, 0. , 0. , 3.21887582,
3.04452244, 1.38629436, 1.60943791, 3.97029191, 1.60943791,
1.79175947, 0. , 0. , 1.09861229, 1.09861229,
1.09861229, 1.09861229, 2.07944154, 4.11087386, 2.63905733,
4.27666612, 0.69314718, 2.30258509, 1.60943791, 0.69314718,
2.63905733, 2.63905733, 0. , 0. , 1.38629436,
3.13549422, 0.69314718, 4.12713439, 1.79175947, 3.04452244,
4.11087386, 3.40119738, 3.98898405, 1.38629436, 0.69314718,
3.17805383, 1.60943791, 1.79175947, 4.55387689, 0. ,
2.30258509, 0. , 0.69314718, 3.8501476 , 1.60943791,
0.69314718, 3.55534806, 0.69314718, 2.48490665, 0.69314718,
4.88280192, 1.09861229, 1.79175947, 0.69314718, 0. ,
2.30258509, 0. , 0. , 0. , 2.94443898,
1.09861229, 0. , 0. , 2.48490665, 1.09861229,
1.09861229, 0. , 0. , 0. , 0.69314718,
0.69314718, 1.60943791, 1.38629436, 1.79175947, 2.07944154,
0. , 1.60943791, 4.33073334, 1.38629436, 1.09861229,
1.79175947, 2.89037176, 2.07944154, 2.7080502 , 3.25809654,
3.63758616, 1.60943791, 0. , 3.04452244, 0.69314718,
3.97029191, 1.09861229, 0. , 1.60943791, 1.38629436,
1.38629436, 2.19722458, 7.94307272, 3.36729583, 0.69314718,
2.7080502 , 3.33220451, 3.80666249, 3.63758616, 2.19722458,
3.58351894, 2.99573227, 1.60943791, 0. , 3.52636052,
1.79175947, 2.30258509, 0. , 0. , 1.94591015,
2.39789527, 0.69314718, 1.38629436, 2.30258509, 0. ,
1.38629436, 0. , 1.94591015, 3.55534806, 2.07944154,
1.38629436, 1.38629436, 3.25809654, 2.63905733, 1.09861229,
6.4035742 , 3.68887945, 1.60943791, 3.09104245, 4.20469262,
0. , 0.69314718, 1.09861229, 0. , 1.09861229,
0.69314718, 1.09861229, 2.07944154, 1.38629436, 1.38629436,
1.60943791, 1.38629436, 0. , 1.38629436, 2.7080502 ,
0. , 2.77258872, 3.17805383, 4.34380542, 3.40119738,
0. , 2.39789527, 1.94591015, 5.04985601, 0. ,
2.19722458, 0.69314718, 0.69314718, 0.69314718, 1.38629436,
2.39789527, 1.94591015, 0. , 1.94591015, 0. ,
0. , 1.09861229, 1.79175947, 2.07944154, 2.19722458,
1.79175947, 0.69314718, 1.79175947, 3.17805383, 6.28413416,
4.73619845, 0.69314718, 1.94591015, 1.94591015, 0. ]),
array([0. , 3.58351894, 2.56494936, 1.60943791, 3.33220451,
3.71357207, 0.69314718, 2.39789527, 3.13549422, 3.17805383,
3.25809654, 0.69314718, 0. , 0. , 3.61091791,
1.60943791, 1.09861229, 0.69314718, 1.94591015, 0.69314718,
3.61091791, 0. , 0. , 1.38629436, 0.69314718,
1.79175947, 2.83321334, 0.69314718, 2.48490665, 0.69314718,
1.38629436, 0. , 1.79175947, 1.60943791, 3.17805383,
1.79175947, 1.09861229, 0.69314718, 0. , 1.38629436,
1.79175947, 0.69314718, 0. , 2.07944154, 2.48490665,
3.04452244, 3.4339872 , 2.7080502 , 1.79175947, 0.69314718,
0. , 1.60943791, 1.60943791, 2.07944154, 0. ,
0. , 2.07944154, 0. , 1.94591015, 1.38629436,
0.69314718, 1.38629436, 0. , 2.30258509, 0.69314718,
0.69314718, 0.69314718, 2.94443898, 2.94443898, 0. ,
0. , 2.63905733, 0. , 0. , 2.83321334,
1.79175947, 0. , 3.04452244, 1.79175947, 0. ,
2.77258872, 1.38629436, 0. , 3.13549422, 0.69314718,
0. , 2.07944154, 1.09861229, 0.69314718, 2.30258509,
0. , 0.69314718, 3.09104245, 2.07944154, 0. ,
0. , 1.60943791, 2.7080502 , 0.69314718, 2.30258509,
2.39789527, 0. , 0. , 2.19722458, 0.69314718,
2.77258872, 1.94591015, 0. , 1.60943791, 1.09861229,
0. , 0. , 1.79175947, 3.13549422, 3.52636052,
1.38629436, 3.66356165, 1.60943791, 4.47733681, 1.79175947,
1.60943791, 3.04452244, 0. , 2.07944154, 2.56494936,
1.38629436, 0.69314718, 0. , 1.79175947, 1.09861229,
1.94591015, 2.77258872, 1.38629436, 0. , 0. ,
1.09861229, 0. , 1.09861229, 2.48490665, 2.19722458,
1.38629436, 2.39789527, 3.13549422, 1.38629436, 1.09861229,
2.48490665, 1.94591015, 1.09861229, 1.79175947, 3.40119738,
1.38629436, 0.69314718, 1.38629436, 3.21887582, 2.89037176,
0. , 1.38629436, 2.56494936, 0.69314718, 0.69314718,
1.09861229, 2.7080502 , 0. , 1.60943791, 3.8501476 ,
1.09861229, 2.48490665, 0. , 1.60943791, 1.79175947,
0. , 2.56494936, 1.38629436, 1.09861229, 0. ,
2.56494936, 0. , 0. , 0.69314718, 1.09861229,
1.09861229, 1.94591015, 0. , 1.94591015, 0. ,
0. , 0. , 0.69314718, 1.09861229, 1.94591015,
1.94591015, 2.48490665, 2.07944154, 3.04452244, 2.07944154,
1.38629436, 0.69314718, 1.38629436, 3.13549422, 0. ]),
array([ 9.77956697, 10.31731758, 7.60787807, 9.73133412, 7.5963923 ,
7.29369772, 7.48885296, 0.69314718, 8.45212119, 3.4339872 ,
2.19722458, 8.61866616, 9.82146372, 6.5971457 , 8.52178264,
13.13692484, 9.22975077, 7.69666708, 3.61091791, 1.60943791,
12.3872352 , 7.3607399 , 7.28961052, 7.43897159, 7.98616486,
2.94443898, 9.57533065, 8.13534695, 2.48490665, 8.05547514,
5.60211882, 1.09861229, 3.09104245, 0.69314718, 8.06148687,
8.50512061, 10.99986401, 9.7251381 , 7.39817409, 6.93439721,
9.24232342, 6.51767127, 8.10681604, 7.12205988, 1.60943791,
9.38117959, 4.72738782, 7.31322039, 9.76457025, 8.04462628,
10.0683662 , 9.51259081, 0. , 5.02388052, 0. ,
9.79300282, 2.39789527, 8.20740183, 6.95368421, 11.22806607,
0. , 0. , 8.96380024, 3.21887582, 8.91958692,
0. , 7.33106031, 6.16541785, 9.2865604 , 6.27098843,
0. , 9.25607826, 10.23113525, 8.22496748, 5.18738581,
8.79694389, 10.27890574, 9.98322252, 8.37239861, 11.05962948,
7.90581031, 7.45760929, 5.88053299, 3.63758616, 10.10699966,
9.51782507, 7.35819375, 0. , 7.46851327, 5.59842196,
9.57831128, 12.3676216 , 5.50938834, 9.59886276, 8.37054761,
9.37067191, 7.32514896, 8.23004431, 4.77912349, 7.24208236,
2.77258872, 9.35314117, 4.53259949, 1.79175947, 5.60211882,
5.95064255, 7.42892719, 7.28344823, 9.41295463, 5.65599181,
8.53542596, 10.46891499, 1.09861229, 2.48490665, 9.37373392,
8.00936308, 7.31322039, 9.4045905 , 5.25227343, 10.15272761,
5.25227343, 0.69314718, 3.36729583, 9.2444519 , 4.44265126,
9.09537835, 9.93081085, 7.3395377 , 7.56060116, 4.26267988,
9.36443391, 6.30809844, 7.52185925, 8.76904081, 9.93532531,
12.63146129, 0. , 2.19722458, 7.48717369, 8.41227702,
8.66198594, 3.8918203 , 10.58215541, 8.63887971, 7.49331725,
1.09861229, 3.8918203 , 1.79175947, 7.96067261, 7.98650494,
5.88053299, 8.43814998, 9.24261393, 7.52510075, 2.07944154,
8.66250492, 8.12563099, 8.01035959, 4.15888308, 6.33505425,
9.06935268, 5.86363118, 7.24351297, 9.79768249, 8.76826315,
9.58169701, 8.32093497, 10.64601996, 0. , 3.71357207,
3.8918203 , 5.70378247, 1.79175947, 0.69314718, 12.58808146,
10.58009876, 4.53259949, 9.85686753, 11.28503316, 8.05547514,
10.63842426, 6.77193556, 10.67373465, 11.72826263, 5.70711026,
11.63407173, 8.05038445, 7.28961052, 7.97625194, 4.83628191,
8.77369415, 7.65917137, 8.74655735, 5.00394631, 11.03438954,
8.01862547, 7.38274645, 0. , 5.31811999, 4.93447393]),
array([0. , 5.51342875, 5.10594547, 5.30330491, 4.96284463,
4.12713439, 4.18965474, 1.38629436, 5.38449506, 2.48490665,
3.29583687, 0.69314718, 6.13556489, 1.09861229, 6.0799332 ,
0. , 0. , 0. , 3.61091791, 0.69314718,
0. , 0. , 1.60943791, 3.09104245, 3.68887945,
1.79175947, 4.38202663, 4.65396035, 2.07944154, 5.09986643,
2.99573227, 0. , 1.94591015, 0. , 4.99721227,
4.94875989, 5.68017261, 1.09861229, 4.18965474, 1.38629436,
0. , 0. , 0. , 2.63905733, 2.48490665,
3.04452244, 3.4339872 , 3.04452244, 0. , 4.29045944,
0. , 0. , 0. , 3.66356165, 0. ,
4.09434456, 1.94591015, 3.76120012, 4.72738782, 0. ,
0. , 0. , 4.78749174, 2.30258509, 2.30258509,
0.69314718, 1.38629436, 4.15888308, 0. , 2.77258872,
0. , 5.14166356, 0. , 3.04452244, 3.49650756,
4.65396035, 0. , 6.94119006, 4.12713439, 5.98896142,
5.35185813, 3.04452244, 0. , 3.13549422, 3.78418963,
0. , 4.26267988, 0. , 3.63758616, 2.48490665,
5.79605775, 0. , 4.24849524, 3.66356165, 0. ,
6.19644413, 4.47733681, 6.47234629, 0.69314718, 3.68887945,
0. , 0. , 0. , 3.13549422, 5.14166356,
0. , 6.31173481, 0. , 0. , 1.79175947,
0. , 6.74993119, 1.38629436, 3.21887582, 0. ,
2.77258872, 4.30406509, 0. , 4.51085951, 0. ,
4.4543473 , 3.04452244, 0. , 4.21950771, 4.15888308,
4.2341065 , 0. , 4.75359019, 3.13549422, 2.39789527,
0. , 3.21887582, 4.57471098, 4.91998093, 0. ,
0. , 0. , 2.30258509, 3.93182563, 5.22035583,
0. , 2.83321334, 4.20469262, 3.91202301, 1.60943791,
2.48490665, 2.7080502 , 1.38629436, 0. , 4.47733681,
0. , 3.98898405, 6.47543272, 3.61091791, 1.60943791,
5.10594547, 0. , 4.84418709, 0.69314718, 0. ,
2.48490665, 3.93182563, 0. , 3.49650756, 6.79009724,
6.79794041, 4.83628191, 0. , 1.60943791, 1.79175947,
2.89037176, 0. , 1.38629436, 0. , 0. ,
7.50714108, 0. , 2.48490665, 0. , 3.09104245,
4.44265126, 3.4657359 , 6.52356231, 0. , 3.25809654,
7.65302041, 3.87120101, 1.94591015, 1.09861229, 4.81218436,
4.54329478, 4.38202663, 0. , 0. , 0. ,
6.35610766, 2.56494936, 0. , 5.0369526 , 0. ]),
array([0. , 0.69314718, 1.94591015, 1.79175947, 1.60943791,
1.60943791, 2.7080502 , 3.52636052, 2.30258509, 1.09861229,
1.09861229, 3.73766962, 3.17805383, 3.13549422, 3.25809654,
0. , 0. , 0. , 3.4339872 , 1.09861229,
0. , 0. , 1.79175947, 1.94591015, 2.89037176,
2.30258509, 1.09861229, 2.39789527, 2.30258509, 0.69314718,
0.69314718, 1.60943791, 3.55534806, 2.39789527, 4.14313473,
0.69314718, 0. , 0.69314718, 3.21887582, 4.48863637,
0. , 0. , 0. , 2.30258509, 3.21887582,
0.69314718, 3.33220451, 2.48490665, 0. , 1.38629436,
0. , 0. , 0. , 1.09861229, 0. ,
1.09861229, 3.40119738, 1.09861229, 1.79175947, 0. ,
0. , 0. , 0.69314718, 4.7095302 , 2.30258509,
2.7080502 , 2.30258509, 2.77258872, 0. , 2.19722458,
0. , 1.60943791, 0. , 0.69314718, 3.55534806,
0.69314718, 0. , 2.30258509, 1.60943791, 0. ,
1.79175947, 0.69314718, 0. , 1.60943791, 1.60943791,
0.69314718, 1.60943791, 0. , 0.69314718, 2.56494936,
2.77258872, 0. , 0.69314718, 1.60943791, 2.19722458,
2.39789527, 0.69314718, 0. , 1.60943791, 1.79175947,
0. , 0. , 0. , 5.15329159, 0.69314718,
0. , 1.79175947, 0. , 0. , 1.94591015,
1.94591015, 1.09861229, 3.29583687, 1.79175947, 0. ,
1.94591015, 2.7080502 , 0. , 1.38629436, 0. ,
3.78418963, 1.09861229, 0.69314718, 0.69314718, 2.19722458,
1.09861229, 0. , 2.89037176, 3.93182563, 5.5174529 ,
0. , 2.56494936, 2.56494936, 1.09861229, 0. ,
0. , 0. , 0.69314718, 0.69314718, 1.09861229,
0. , 2.48490665, 0.69314718, 1.38629436, 0.69314718,
1.09861229, 2.63905733, 3.25809654, 0. , 4.24849524,
0. , 2.89037176, 2.48490665, 2.39789527, 1.60943791,
1.79175947, 0. , 1.79175947, 3.09104245, 0. ,
0.69314718, 2.19722458, 0.69314718, 2.48490665, 2.56494936,
1.38629436, 1.60943791, 0. , 0.69314718, 0.69314718,
3.04452244, 0. , 3.25809654, 2.19722458, 0. ,
0. , 1.79175947, 2.48490665, 0. , 1.60943791,
0. , 3.63758616, 0. , 0. , 1.09861229,
0. , 0.69314718, 4.26267988, 5.71042702, 1.79175947,
2.99573227, 1.94591015, 0. , 0. , 0. ,
1.09861229, 0.69314718, 2.99573227, 4.39444915, 0.69314718]),
array([3.52636052, 3.58351894, 3.58351894, 1.60943791, 3.49650756,
3.63758616, 3.25809654, 3.87120101, 3.58351894, 2.19722458,
3.33220451, 0.69314718, 1.60943791, 1.09861229, 3.49650756,
1.60943791, 3.49650756, 0.69314718, 3.21887582, 0.69314718,
0. , 2.89037176, 1.38629436, 2.39789527, 1.38629436,
1.79175947, 2.83321334, 1.09861229, 3.4339872 , 0.69314718,
1.94591015, 0. , 1.79175947, 1.60943791, 0. ,
3.13549422, 1.09861229, 0.69314718, 0.69314718, 1.38629436,
1.79175947, 1.60943791, 0.69314718, 2.77258872, 2.7080502 ,
2.99573227, 3.52636052, 2.99573227, 2.19722458, 0.69314718,
1.09861229, 2.07944154, 1.60943791, 3.4339872 , 0. ,
0. , 3.73766962, 3.4339872 , 3.78418963, 1.38629436,
0.69314718, 1.38629436, 2.99573227, 2.30258509, 0.69314718,
2.39789527, 1.09861229, 1.60943791, 0.69314718, 1.09861229,
2.07944154, 2.63905733, 1.38629436, 1.38629436, 3.55534806,
2.89037176, 0. , 0.69314718, 2.7080502 , 2.07944154,
3.78418963, 2.7080502 , 2.39789527, 0.69314718, 2.56494936,
0. , 3.76120012, 1.38629436, 2.19722458, 2.77258872,
0. , 1.38629436, 3.52636052, 2.83321334, 0. ,
2.63905733, 1.79175947, 3.63758616, 0.69314718, 3.49650756,
2.39789527, 0. , 0. , 3.09104245, 2.99573227,
3.09104245, 3.8918203 , 1.94591015, 2.99573227, 2.19722458,
0. , 0. , 3.91202301, 3.13549422, 0.69314718,
1.79175947, 3.58351894, 1.60943791, 3.91202301, 1.38629436,
2.94443898, 2.7080502 , 0. , 2.83321334, 2.99573227,
3.76120012, 2.56494936, 3.04452244, 1.79175947, 0.69314718,
2.39789527, 2.94443898, 1.94591015, 1.94591015, 1.79175947,
1.09861229, 0. , 1.09861229, 3.25809654, 1.60943791,
1.38629436, 2.39789527, 3.36729583, 1.38629436, 1.09861229,
2.48490665, 3.4339872 , 2.99573227, 3.4657359 , 3.58351894,
1.60943791, 2.83321334, 1.38629436, 2.30258509, 2.94443898,
1.94591015, 1.79175947, 3.73766962, 1.09861229, 1.79175947,
1.38629436, 3.58351894, 0. , 2.30258509, 2.30258509,
2.07944154, 3.58351894, 0. , 1.60943791, 1.79175947,
2.39789527, 3.17805383, 2.48490665, 2.07944154, 2.07944154,
2.99573227, 0. , 0. , 0.69314718, 1.38629436,
1.79175947, 3.76120012, 1.09861229, 1.79175947, 3.25809654,
1.09861229, 0. , 1.94591015, 1.38629436, 3.55534806,
2.07944154, 1.09861229, 3.76120012, 3.04452244, 2.07944154,
1.38629436, 2.56494936, 1.38629436, 2.94443898, 0. ]),
array([0. , 5.08140436, 6.26149168, 3.40119738, 3.63758616,
3.8501476 , 6.19440539, 4.84418709, 4.91265489, 3.17805383,
3.25809654, 3.52636052, 5.44673737, 1.09861229, 5.26269019,
1.60943791, 4.02535169, 0.69314718, 5.64190707, 0.69314718,
4.82028157, 4.02535169, 1.60943791, 3.09104245, 6.08221891,
1.79175947, 4.6443909 , 1.94591015, 3.68887945, 3.49650756,
2.99573227, 0. , 1.79175947, 1.60943791, 6.62539237,
4.12713439, 4.47733681, 0.69314718, 4.73619845, 4.88280192,
2.63905733, 1.60943791, 0. , 4.49980967, 4.24849524,
3.04452244, 4.97673374, 4.59511985, 2.19722458, 3.8918203 ,
3.87120101, 3.58351894, 1.60943791, 3.29583687, 0. ,
4.59511985, 5.66642669, 3.63758616, 4.71849887, 1.38629436,
0.69314718, 1.38629436, 3.4339872 , 5.00394631, 2.30258509,
3.71357207, 1.09861229, 6.68959927, 5.64190707, 6.37331979,
2.83321334, 4.57471098, 0. , 3.04452244, 5.30330491,
4.31748811, 0. , 5.02388052, 4.2341065 , 4.81218436,
5.49306144, 2.83321334, 2.48490665, 4.91265489, 4.14313473,
0. , 4.77068462, 2.63905733, 2.48490665, 6.3526294 ,
6.07073773, 0.69314718, 3.52636052, 3.25809654, 0. ,
4.94875989, 2.48490665, 6.07073773, 3.49650756, 3.63758616,
2.39789527, 0.69314718, 0. , 6.34388043, 5.28320373,
3.09104245, 6.02344759, 2.39789527, 4.30406509, 3.4339872 ,
1.38629436, 4.59511985, 4.36944785, 3.13549422, 6.22059017,
4.99043259, 5.59842196, 1.60943791, 4.48863637, 1.79175947,
5.18178355, 3.04452244, 0. , 4.21950771, 2.94443898,
5.12989871, 0. , 5.62401751, 4.59511985, 6.94215671,
1.94591015, 3.04452244, 2.48490665, 4.59511985, 2.19722458,
4.66343909, 0. , 1.09861229, 4.63472899, 5.22035583,
2.30258509, 5.09986643, 4.11087386, 4.09434456, 1.09861229,
2.48490665, 4.2341065 , 4.73619845, 0.69314718, 4.41884061,
1.79175947, 5.6094718 , 4.85203026, 5.170484 , 2.94443898,
5.08140436, 1.60943791, 4.80402104, 4.69134788, 1.38629436,
2.07944154, 3.63758616, 0. , 4.59511985, 6.76041469,
4.07753744, 4.71849887, 0. , 1.60943791, 1.79175947,
4.44265126, 3.29583687, 6.58340922, 3.8286414 , 0. ,
5.7651911 , 0. , 4.49980967, 0. , 3.21887582,
3.91202301, 5.81413053, 3.04452244, 2.99573227, 4.2341065 ,
4.56434819, 0.69314718, 2.83321334, 4.00733319, 4.68213123,
3.63758616, 4.74493213, 0.69314718, 3.04452244, 1.09861229,
5.37063803, 2.77258872, 1.38629436, 6.23244802, 0. ]),
array([0. , 0. , 0. , 0. , 0. ,
0. , 0. , 1.38629436, 0. , 0.69314718,
0.69314718, 0. , 0. , 0.69314718, 0. ,
0. , 0. , 0. , 0.69314718, 0.69314718,
0. , 0. , 0. , 0. , 0. ,
0.69314718, 0. , 0. , 0.69314718, 0. ,
0.69314718, 0.69314718, 0.69314718, 0.69314718, 0. ,
0. , 0. , 0. , 0. , 0.69314718,
0. , 0. , 0. , 0. , 0.69314718,
0. , 1.09861229, 0. , 0. , 0. ,
0. , 0. , 0. , 0.69314718, 0. ,
0. , 0.69314718, 0. , 0.69314718, 0. ,
0. , 0. , 0. , 0.69314718, 0. ,
0.69314718, 0. , 0.69314718, 0. , 0.69314718,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.69314718, 0. ,
0. , 0. , 0. , 0. , 0.69314718,
0. , 0. , 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0.69314718, 0. ,
0. , 0. , 0. , 0.69314718, 0. ,
0. , 0. , 0. , 0. , 0.69314718,
0. , 0. , 0. , 1.38629436, 0. ,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 0.69314718, 0.69314718, 0. , 0.69314718,
0. , 0. , 0. , 0. , 0.69314718,
0. , 1.09861229, 0. , 0. , 0. ,
0. , 0. , 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.69314718, 1.09861229, 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0. , 0.69314718,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 0.69314718, 0. , 0. , 0. ,
0. , 0. , 0. , 0.69314718, 0.69314718,
0.69314718, 0. , 0.69314718, 0.69314718, 0. ,
0. , 0.69314718, 0. , 0. , 0. ,
0. , 0.69314718, 0. , 0. , 0.69314718,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.69314718, 0. , 0.69314718]),
array([0. , 0.69314718, 1.38629436, 1.09861229, 0.69314718,
0.69314718, 2.07944154, 1.79175947, 0.69314718, 0.69314718,
0.69314718, 1.79175947, 1.09861229, 1.60943791, 2.30258509,
0. , 0. , 0. , 1.79175947, 0.69314718,
0. , 0. , 0.69314718, 0.69314718, 2.07944154,
0.69314718, 0.69314718, 0.69314718, 1.09861229, 0.69314718,
0.69314718, 1.09861229, 1.60943791, 0.69314718, 2.30258509,
0.69314718, 0. , 0.69314718, 1.94591015, 2.77258872,
0. , 0. , 0. , 1.38629436, 2.07944154,
0.69314718, 1.60943791, 1.38629436, 0. , 0.69314718,
0. , 0. , 0. , 0.69314718, 0. ,
0.69314718, 1.60943791, 0.69314718, 0.69314718, 0. ,
0. , 0. , 0.69314718, 2.39789527, 0.69314718,
1.38629436, 0.69314718, 1.60943791, 0. , 1.60943791,
0. , 0.69314718, 0. , 0.69314718, 2.30258509,
0.69314718, 0. , 1.09861229, 1.09861229, 0. ,
1.09861229, 0.69314718, 0. , 1.38629436, 0.69314718,
0.69314718, 0.69314718, 0. , 0.69314718, 2.07944154,
2.30258509, 0. , 0.69314718, 1.38629436, 0.69314718,
0.69314718, 0.69314718, 0. , 0.69314718, 1.09861229,
0. , 0. , 0. , 3.09104245, 0.69314718,
0. , 0.69314718, 0. , 0. , 1.60943791,
1.79175947, 0.69314718, 0.69314718, 1.38629436, 0. ,
1.38629436, 1.60943791, 0. , 1.09861229, 0. ,
1.38629436, 0.69314718, 0.69314718, 0.69314718, 1.09861229,
1.09861229, 0. , 1.79175947, 1.60943791, 3.49650756,
0. , 1.09861229, 0.69314718, 0.69314718, 0. ,
0. , 0. , 0.69314718, 0.69314718, 0.69314718,
0. , 1.38629436, 0. , 0.69314718, 0.69314718,
1.09861229, 1.60943791, 1.09861229, 0. , 1.79175947,
0. , 0.69314718, 1.38629436, 1.09861229, 0.69314718,
1.79175947, 0. , 1.09861229, 2.30258509, 0. ,
0.69314718, 0.69314718, 0.69314718, 2.39789527, 0.69314718,
0.69314718, 0.69314718, 0. , 0.69314718, 0.69314718,
1.09861229, 0. , 1.94591015, 2.07944154, 0. ,
0. , 0.69314718, 1.09861229, 0. , 1.09861229,
0. , 2.83321334, 0. , 0. , 0.69314718,
0. , 0.69314718, 2.39789527, 2.77258872, 1.38629436,
0.69314718, 1.38629436, 0. , 0. , 0. ,
0.69314718, 0.69314718, 2.30258509, 2.48490665, 0.69314718]),
array([0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0. , 0.69314718,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 2.07944154, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.69314718, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.69314718, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 1.09861229, 0. , 0. ,
0. , 0. , 0. , 0. , 1.38629436,
0. , 0. , 0. , 1.09861229, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.69314718, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.69314718, 0.69314718, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ]),
array([0. , 3.58351894, 2.94443898, 1.60943791, 3.33220451,
3.71357207, 3.25809654, 1.09861229, 3.78418963, 2.48490665,
3.25809654, 0.69314718, 0. , 0. , 3.61091791,
1.60943791, 1.09861229, 0.69314718, 1.94591015, 0.69314718,
4.14313473, 0. , 0. , 2.07944154, 0.69314718,
1.79175947, 2.83321334, 0.69314718, 2.07944154, 0.69314718,
2.99573227, 0. , 1.79175947, 0. , 3.68887945,
2.48490665, 1.09861229, 0.69314718, 0. , 1.38629436,
1.79175947, 0.69314718, 0. , 2.07944154, 2.48490665,
3.04452244, 3.4339872 , 2.7080502 , 1.94591015, 0.69314718,
1.09861229, 1.60943791, 0.69314718, 3.25809654, 0. ,
0. , 1.79175947, 0. , 1.94591015, 1.38629436,
0.69314718, 1.38629436, 0. , 2.30258509, 0.69314718,
0. , 0.69314718, 4.15888308, 2.94443898, 0. ,
2.07944154, 2.63905733, 0. , 0. , 2.83321334,
1.79175947, 0. , 3.78418963, 1.79175947, 0.69314718,
2.77258872, 1.38629436, 0. , 3.13549422, 0.69314718,
0. , 3.61091791, 1.09861229, 0.69314718, 2.30258509,
0. , 0.69314718, 3.4657359 , 2.89037176, 0. ,
0.69314718, 1.60943791, 2.7080502 , 0.69314718, 2.39789527,
2.39789527, 0. , 0. , 0. , 0.69314718,
2.77258872, 4.63472899, 1.09861229, 1.60943791, 1.79175947,
0. , 0. , 1.38629436, 3.09104245, 3.52636052,
1.38629436, 3.66356165, 1.60943791, 4.48863637, 1.79175947,
2.19722458, 2.56494936, 0. , 2.07944154, 2.94443898,
1.38629436, 1.60943791, 0. , 1.79175947, 1.09861229,
1.94591015, 2.83321334, 1.38629436, 1.09861229, 0. ,
1.09861229, 0. , 1.09861229, 2.48490665, 3.8501476 ,
1.38629436, 1.94591015, 3.40119738, 1.38629436, 1.09861229,
2.48490665, 1.94591015, 1.09861229, 3.4657359 , 3.49650756,
1.38629436, 0.69314718, 1.38629436, 3.21887582, 1.38629436,
1.79175947, 1.38629436, 2.56494936, 0.69314718, 0.69314718,
1.09861229, 2.94443898, 0. , 1.60943791, 3.8501476 ,
1.09861229, 2.7080502 , 0. , 1.60943791, 1.79175947,
0. , 3.29583687, 1.38629436, 0. , 0. ,
2.56494936, 0. , 0. , 0.69314718, 1.09861229,
1.09861229, 1.94591015, 0. , 1.94591015, 0. ,
0. , 0. , 1.38629436, 1.09861229, 3.21887582,
2.07944154, 2.48490665, 3.58351894, 3.04452244, 2.07944154,
1.38629436, 2.56494936, 0. , 3.13549422, 0. ])]
train_model_input
{'I1': 0 0.000000
1 0.000000
2 0.000000
3 0.000000
4 0.000000
...
195 0.000000
196 0.000000
197 0.693147
198 0.000000
199 0.693147
Name: I1, Length: 200, dtype: float64,
'I2': 0 1.386294
1 -1.000000
2 0.000000
3 2.639057
4 0.000000
...
195 0.000000
196 0.693147
197 0.000000
198 3.135494
199 -1.000000
Name: I2, Length: 200, dtype: float64,
'I3': 0 5.564520
1 2.995732
2 1.098612
3 0.693147
4 4.653960
...
195 4.736198
196 0.693147
197 1.945910
198 1.945910
199 0.000000
Name: I3, Length: 200, dtype: float64,
'I4': 0 0.000000
1 3.583519
2 2.564949
3 1.609438
4 3.332205
...
195 1.386294
196 0.693147
197 1.386294
198 3.135494
199 0.000000
Name: I4, Length: 200, dtype: float64,
'I5': 0 9.779567
1 10.317318
2 7.607878
3 9.731334
4 7.596392
...
195 8.018625
196 7.382746
197 0.000000
198 5.318120
199 4.934474
Name: I5, Length: 200, dtype: float64,
'I6': 0 0.000000
1 5.513429
2 5.105945
3 5.303305
4 4.962845
...
195 6.356108
196 2.564949
197 0.000000
198 5.036953
199 0.000000
Name: I6, Length: 200, dtype: float64,
'I7': 0 0.000000
1 0.693147
2 1.945910
3 1.791759
4 1.609438
...
195 1.098612
196 0.693147
197 2.995732
198 4.394449
199 0.693147
Name: I7, Length: 200, dtype: float64,
'I8': 0 3.526361
1 3.583519
2 3.583519
3 1.609438
4 3.496508
...
195 1.386294
196 2.564949
197 1.386294
198 2.944439
199 0.000000
Name: I8, Length: 200, dtype: float64,
'I9': 0 0.000000
1 5.081404
2 6.261492
3 3.401197
4 3.637586
...
195 5.370638
196 2.772589
197 1.386294
198 6.232448
199 0.000000
Name: I9, Length: 200, dtype: float64,
'I10': 0 0.000000
1 0.000000
2 0.000000
3 0.000000
4 0.000000
...
195 0.000000
196 0.000000
197 0.693147
198 0.000000
199 0.693147
Name: I10, Length: 200, dtype: float64,
'I11': 0 0.000000
1 0.693147
2 1.386294
3 1.098612
4 0.693147
...
195 0.693147
196 0.693147
197 2.302585
198 2.484907
199 0.693147
Name: I11, Length: 200, dtype: float64,
'I12': 0 0.0
1 0.0
2 0.0
3 0.0
4 0.0
...
195 0.0
196 0.0
197 0.0
198 0.0
199 0.0
Name: I12, Length: 200, dtype: float64,
'I13': 0 0.000000
1 3.583519
2 2.944439
3 1.609438
4 3.332205
...
195 1.386294
196 2.564949
197 0.000000
198 3.135494
199 0.000000
Name: I13, Length: 200, dtype: float64,
'C1': 0 0
1 11
2 0
3 0
4 0
..
195 0
196 21
197 0
198 0
199 21
Name: C1, Length: 200, dtype: int32,
'C2': 0 4
1 1
2 18
3 45
4 11
..
195 0
196 64
197 5
198 84
199 68
Name: C2, Length: 200, dtype: int32,
'C3': 0 96
1 98
2 39
3 7
4 59
...
195 66
196 44
197 153
198 84
199 0
Name: C3, Length: 200, dtype: int32,
'C4': 0 146
1 98
2 52
3 117
4 77
...
195 107
196 65
197 143
198 41
199 0
Name: C4, Length: 200, dtype: int32,
'C5': 0 1
1 1
2 3
3 1
4 1
..
195 1
196 1
197 1
198 5
199 5
Name: C5, Length: 200, dtype: int32,
'C6': 0 4
1 6
2 4
3 0
4 5
..
195 4
196 6
197 0
198 4
199 0
Name: C6, Length: 200, dtype: int32,
'C7': 0 163
1 179
2 140
3 164
4 18
...
195 10
196 4
197 141
198 27
199 19
Name: C7, Length: 200, dtype: int32,
'C8': 0 1
1 0
2 2
3 1
4 1
..
195 15
196 2
197 1
198 1
199 1
Name: C8, Length: 200, dtype: int32,
'C9': 0 1
1 1
2 1
3 0
4 1
..
195 1
196 1
197 1
198 1
199 1
Name: C9, Length: 200, dtype: int32,
'C10': 0 72
1 89
2 93
3 20
4 45
...
195 11
196 9
197 118
198 96
199 61
Name: C10, Length: 200, dtype: int32,
'C11': 0 117
1 58
2 31
3 61
4 171
...
195 87
196 130
197 30
198 110
199 149
Name: C11, Length: 200, dtype: int32,
'C12': 0 127
1 97
2 122
3 104
4 162
...
195 6
196 96
197 128
198 67
199 0
Name: C12, Length: 200, dtype: int32,
'C13': 0 157
1 79
2 16
3 36
4 96
...
195 73
196 57
197 69
198 63
199 14
Name: C13, Length: 200, dtype: int32,
'C14': 0 7
1 7
2 7
3 1
4 4
..
195 2
196 1
197 13
198 1
199 1
Name: C14, Length: 200, dtype: int32,
'C15': 0 127
1 72
2 129
3 43
4 36
...
195 53
196 106
197 23
198 121
199 24
Name: C15, Length: 200, dtype: int32,
'C16': 0 126
1 26
2 97
3 43
4 121
...
195 49
196 1
197 4
198 75
199 0
Name: C16, Length: 200, dtype: int32,
'C17': 0 8
1 7
2 8
3 8
4 8
..
195 0
196 1
197 4
198 4
199 7
Name: C17, Length: 200, dtype: int32,
'C18': 0 66
1 52
2 49
3 37
4 14
..
195 74
196 25
197 40
198 7
199 72
Name: C18, Length: 200, dtype: int32,
'C19': 0 0
1 0
2 0
3 0
4 5
..
195 5
196 0
197 17
198 18
199 0
Name: C19, Length: 200, dtype: int32,
'C20': 0 0
1 0
2 0
3 0
4 3
..
195 1
196 0
197 2
198 1
199 0
Name: C20, Length: 200, dtype: int32,
'C21': 0 3
1 47
2 25
3 156
4 9
...
195 30
196 138
197 41
198 123
199 0
Name: C21, Length: 200, dtype: int32,
'C22': 0 0
1 0
2 0
3 0
4 0
..
195 5
196 0
197 0
198 0
199 0
Name: C22, Length: 200, dtype: int32,
'C23': 0 1
1 7
2 6
3 0
4 0
..
195 0
196 0
197 0
198 0
199 0
Name: C23, Length: 200, dtype: int32,
'C24': 0 96
1 112
2 53
3 32
4 5
...
195 118
196 68
197 12
198 10
199 0
Name: C24, Length: 200, dtype: int32,
'C25': 0 0
1 0
2 0
3 0
4 1
..
195 17
196 0
197 16
198 16
199 0
Name: C25, Length: 200, dtype: int32,
'C26': 0 0
1 0
2 0
3 0
4 47
..
195 48
196 0
197 11
198 49
199 0
Name: C26, Length: 200, dtype: int32}
from keras.utils import plot_model
plot_model(model, to_file='./DeepCrossing.png',show_shapes=True)
[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-lVocUFRC-1615871597237)(output_49_0.png)]
# 模型训练
model.fit(train_model_input, train_data['label'].values,
batch_size=64, epochs=5, validation_split=0.2, )
Epoch 1/5
3/3 [==============================] - 0s 77ms/step - loss: 0.4399 - auc: 0.7993 - val_loss: 0.5972 - val_auc: 0.7208
Epoch 2/5
3/3 [==============================] - 0s 26ms/step - loss: 0.4174 - auc: 0.8225 - val_loss: 0.5646 - val_auc: 0.7051
Epoch 3/5
3/3 [==============================] - 0s 20ms/step - loss: 0.4097 - auc: 0.8430 - val_loss: 0.5561 - val_auc: 0.6909
Epoch 4/5
3/3 [==============================] - 0s 19ms/step - loss: 0.4073 - auc: 0.8641 - val_loss: 0.5570 - val_auc: 0.6937
Epoch 5/5
3/3 [==============================] - 0s 19ms/step - loss: 0.3897 - auc: 0.8763 - val_loss: 0.5719 - val_auc: 0.7051
<tensorflow.python.keras.callbacks.History at 0x24a506dff40>
# 模型训练
model.fit(train_dense_x+train_sparse_x, train_data['label'].values,
batch_size=64, epochs=5, validation_split=0.2, )
Epoch 1/5
3/3 [==============================] - 0s 60ms/step - loss: 0.3778 - auc: 0.8881 - val_loss: 0.5980 - val_auc: 0.7066
Epoch 2/5
3/3 [==============================] - 0s 22ms/step - loss: 0.3670 - auc: 0.8998 - val_loss: 0.5927 - val_auc: 0.6909
Epoch 3/5
3/3 [==============================] - 0s 20ms/step - loss: 0.3501 - auc: 0.9107 - val_loss: 0.5797 - val_auc: 0.7009
Epoch 4/5
3/3 [==============================] - 0s 23ms/step - loss: 0.3332 - auc: 0.9247 - val_loss: 0.5725 - val_auc: 0.6994
Epoch 5/5
3/3 [==============================] - 0s 20ms/step - loss: 0.3211 - auc: 0.9316 - val_loss: 0.5785 - val_auc: 0.6923
2. 模型结构及原理
为了完成端到端的训练， DeepCrossing模型要在内部网络结构中解决如下问题：
- 离散类特征编码后过于稀疏， 不利于直接输入神经网络训练， 需要解决稀疏特征向量稠密化的问题
- 如何解决特征自动交叉组合的问题
- 如何在输出层中达成问题设定的优化目标
DeepCrossing分别设置了不同神经网络层解决上述问题。模型结构如下
下面分别介绍一下各层的作用：
2.1 Embedding Layer
将稀疏的类别型特征转成稠密的Embedding向量，Embedding的维度会远小于原始的稀疏特征向量。 Embedding是NLP里面常用的一种技术，这里的Feature #1表示的类别特征(one-hot编码后的稀疏特征向量）， Feature #2是数值型特征，不用embedding， 直接到了Stacking Layer。 关于Embedding Layer的实现， 往往一个全连接层即可，Tensorflow中有实现好的层可以直接用。 和NLP里面的embedding技术异曲同工， 比如Word2Vec， 语言模型等。
2.2 Stacking Layer
这个层是把不同的Embedding特征和数值型特征拼接在一起，形成新的包含全部特征的特征向量，该层通常也称为连接层, 具体的实现如下，先将所有的数值特征拼接起来，然后将所有的Embedding拼接起来，最后将数值特征和Embedding特征拼接起来作为DNN的输入，这里TF是通过Concatnate层进行拼接。
2.3 Multiple Residual Units Layer
该层的主要结构是MLP， 但DeepCrossing采用了残差网络进行的连接。通过多层残差网络对特征向量各个维度充分的交叉组合， 使得模型能够抓取更多的非线性特征和组合特征信息， 增加模型的表达能力。残差网络结构如下图所示：
Deep Crossing模型使用稍微修改过的残差单元，它不使用卷积内核，改为了两层神经网络。我们可以看到，残差单元是通过两层ReLU变换再将原输入特征相加回来实现的。具体代码实现如下：
2.4 Scoring Layer
这个作为输出层，为了拟合优化目标存在。 对于CTR预估二分类问题， Scoring往往采用逻辑回归，模型通过叠加多个残差块加深网络的深度，最后将结果转换成一个概率值输出。
200
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
