30天干掉tensorflow2.0-day17 特征列feature_column

特征列feature_column

特征列 通常用于对结构化数据实施特征工程时候使用，图像或者文本数据一般不会用到特征列。

一，特征列用法概述

使用特征列可以将类别特征转换为one-hot编码特征，将连续特征构建分桶特征，以及对多个特征生成交叉特征等等。

要创建特征列，请调用 tf.feature_column 模块的函数。该模块中常用的九个函数如下图所示，所有九个函数都会返回一个 Categorical-Column 或一个
Dense-Column 对象，但却不会返回 bucketized_column，后者继承自这两个类。

注意：所有的Catogorical Column类型最终都要通过indicator_column转换成Dense Column类型才能传入模型！

• numeric_column 数值列，最常用。

• bucketized_column 分桶列，由数值列生成，可以由一个数值列出多个特征，one-hot编码。

• categorical_column_with_identity 分类标识列，one-hot编码，相当于分桶列每个桶为1个整数的情况。

• categorical_column_with_vocabulary_list 分类词汇列，one-hot编码，由list指定词典。

• categorical_column_with_vocabulary_file 分类词汇列，由文件file指定词典。

• categorical_column_with_hash_bucket 哈希列，整数或词典较大时采用。

• indicator_column 指标列，由Categorical Column生成，one-hot编码

• embedding_column 嵌入列，由Categorical Column生成，嵌入矢量分布参数需要学习。嵌入矢量维数建议取类别数量的 4 次方根。

• crossed_column 交叉列，可以由除categorical_column_with_hash_bucket的任意分类列构成。

二，特征列使用范例

以下是一个使用特征列解决Titanic生存问题的完整范例。

import datetime
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
import tensorflow as tf
from tensorflow.keras import layers,models


#打印日志
def printlog(info):
    nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
    print("\n"+"=========="*8 + "%s"%nowtime)
    print(info+'...\n\n')

#================================================================================
# 一，构建数据管道
#================================================================================
printlog("step1: prepare dataset...")


dftrain_raw = pd.read_csv("./data/titanic/train.csv")
dftest_raw = pd.read_csv("./data/titanic/test.csv")

dfraw = pd.concat([dftrain_raw,dftest_raw])

def prepare_dfdata(dfraw):
    dfdata = dfraw.copy()
    dfdata.columns = [x.lower() for x in dfdata.columns]
    dfdata = dfdata.rename(columns={'survived':'label'})
    dfdata = dfdata.drop(['passengerid','name'],axis = 1)
    for col,dtype in dict(dfdata.dtypes).items():
        # 判断是否包含缺失值
        if dfdata[col].hasnans:
            # 添加标识是否缺失列
            dfdata[col + '_nan'] = pd.isna(dfdata[col]).astype('int32')
            # 填充
            if dtype not in [np.object,np.str,np.unicode]:
                dfdata[col].fillna(dfdata[col].mean(),inplace = True)
            else:
                dfdata[col].fillna('',inplace = True)
    return(dfdata)

dfdata = prepare_dfdata(dfraw)
dftrain = dfdata.iloc[0:len(dftrain_raw),:]
dftest = dfdata.iloc[len(dftrain_raw):,:]



# 从 dataframe 导入数据 
def df_to_dataset(df, shuffle=True, batch_size=32):
    dfdata = df.copy()
    if 'label' not in dfdata.columns:
        ds = tf.data.Dataset.from_tensor_slices(dfdata.to_dict(orient = 'list'))
    else: 
        labels = dfdata.pop('label')
        ds = tf.data.Dataset.from_tensor_slices((dfdata.to_dict(orient = 'list'), labels))  
    if shuffle:
        ds = ds.shuffle(buffer_size=len(dfdata))
    ds = ds.batch(batch_size)
    return ds

ds_train = df_to_dataset(dftrain)
ds_test = df_to_dataset(dftest)

================================================================================2020-04-21 08:16:42
step1: prepare dataset......

#================================================================================
# 二，定义特征列
#================================================================================
printlog("step2: make feature columns...")

feature_columns = []

# 数值列
for col in ['age','fare','parch','sibsp'] + [
    c for c in dfdata.columns if c.endswith('_nan')]:
    feature_columns.append(tf.feature_column.numeric_column(col))

# 分桶列
age = tf.feature_column.numeric_column('age')
age_buckets = tf.feature_column.bucketized_column(age, 
             boundaries=[18, 25, 30, 35, 40, 45, 50, 55, 60, 65])
feature_columns.append(age_buckets)

# 类别列
# 注意：所有的Catogorical Column类型最终都要通过indicator_column转换成Dense Column类型才能传入模型！！
sex = tf.feature_column.indicator_column(
      tf.feature_column.categorical_column_with_vocabulary_list(
      key='sex',vocabulary_list=["male", "female"]))
feature_columns.append(sex)

pclass = tf.feature_column.indicator_column(
      tf.feature_column.categorical_column_with_vocabulary_list(
      key='pclass',vocabulary_list=[1,2,3]))
feature_columns.append(pclass)

ticket = tf.feature_column.indicator_column(
     tf.feature_column.categorical_column_with_hash_bucket('ticket',3))
feature_columns.append(ticket)

embarked = tf.feature_column.indicator_column(
      tf.feature_column.categorical_column_with_vocabulary_list(
      key='embarked',vocabulary_list=['S','C','B']))
feature_columns.append(embarked)

# 嵌入列
cabin = tf.feature_column.embedding_column(
    tf.feature_column.categorical_column_with_hash_bucket('cabin',32),2)
feature_columns.append(cabin)

# 交叉列
pclass_cate = tf.feature_column.categorical_column_with_vocabulary_list(
          key='pclass',vocabulary_list=[1,2,3])

crossed_feature = tf.feature_column.indicator_column(
    tf.feature_column.crossed_column([age_buckets, pclass_cate],hash_bucket_size=15))

feature_columns.append(crossed_feature)

================================================================================2020-04-21 08:16:57
step2: make feature columns......

#================================================================================
# 三，定义模型
#================================================================================
printlog("step3: define model...")

tf.keras.backend.clear_session()
model = tf.keras.Sequential([
  layers.DenseFeatures(feature_columns), #将特征列放入到tf.keras.layers.DenseFeatures中!!!
  layers.Dense(64, activation='relu'),
  layers.Dense(64, activation='relu'),
  layers.Dense(1, activation='sigmoid')
])

================================================================================2020-04-21 08:17:10
step3: define model......

#================================================================================
# 四，训练模型
#================================================================================
printlog("step4: train model...")

model.compile(optimizer='adam',
              loss='binary_crossentropy',
              metrics=['accuracy'])

history = model.fit(ds_train,
          validation_data=ds_test,
          epochs=10)

================================================================================2020-04-21 08:17:16
step4: train model......


WARNING:tensorflow:From D:\anaconda3\lib\site-packages\tensorflow_core\python\feature_column\feature_column_v2.py:4267: IndicatorColumn._variable_shape (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead.
WARNING:tensorflow:From D:\anaconda3\lib\site-packages\tensorflow_core\python\feature_column\feature_column_v2.py:4322: CrossedColumn._num_buckets (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead.
WARNING:tensorflow:From D:\anaconda3\lib\site-packages\tensorflow_core\python\feature_column\feature_column_v2.py:4322: VocabularyListCategoricalColumn._num_buckets (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead.
WARNING:tensorflow:From D:\anaconda3\lib\site-packages\tensorflow_core\python\feature_column\feature_column_v2.py:4322: HashedCategoricalColumn._num_buckets (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead.
Train for 23 steps, validate for 6 steps
Epoch 1/10
23/23 [==============================] - 2s 82ms/step - loss: 1.4773 - accuracy: 0.5702 - val_loss: 0.7977 - val_accuracy: 0.6704
Epoch 2/10
23/23 [==============================] - 0s 7ms/step - loss: 0.7046 - accuracy: 0.6713 - val_loss: 0.6846 - val_accuracy: 0.6872
Epoch 3/10
23/23 [==============================] - 0s 8ms/step - loss: 0.6275 - accuracy: 0.6854 - val_loss: 0.6108 - val_accuracy: 0.6648
Epoch 4/10
23/23 [==============================] - 0s 10ms/step - loss: 0.5921 - accuracy: 0.6882 - val_loss: 0.5754 - val_accuracy: 0.7039
Epoch 5/10
23/23 [==============================] - 0s 9ms/step - loss: 0.5719 - accuracy: 0.7275 - val_loss: 0.5477 - val_accuracy: 0.7207
Epoch 6/10
23/23 [==============================] - 0s 8ms/step - loss: 0.5512 - accuracy: 0.7346 - val_loss: 0.5271 - val_accuracy: 0.7318
Epoch 7/10
23/23 [==============================] - 0s 11ms/step - loss: 0.5449 - accuracy: 0.7416 - val_loss: 0.5693 - val_accuracy: 0.7430
Epoch 8/10
23/23 [==============================] - 0s 10ms/step - loss: 0.5788 - accuracy: 0.7303 - val_loss: 0.5126 - val_accuracy: 0.7263
Epoch 9/10
23/23 [==============================] - 0s 8ms/step - loss: 0.5146 - accuracy: 0.7626 - val_loss: 0.4921 - val_accuracy: 0.7430
Epoch 10/10
23/23 [==============================] - 0s 7ms/step - loss: 0.5041 - accuracy: 0.7697 - val_loss: 0.4957 - val_accuracy: 0.7486

#================================================================================
# 五，评估模型
#================================================================================
printlog("step5: eval model...")

model.summary()


%matplotlib inline
%config InlineBackend.figure_format = 'svg'

import matplotlib.pyplot as plt

def plot_metric(history, metric):
    train_metrics = history.history[metric]
    val_metrics = history.history['val_'+metric]
    epochs = range(1, len(train_metrics) + 1)
    plt.plot(epochs, train_metrics, 'bo--')
    plt.plot(epochs, val_metrics, 'ro-')
    plt.title('Training and validation '+ metric)
    plt.xlabel("Epochs")
    plt.ylabel(metric)
    plt.legend(["train_"+metric, 'val_'+metric])
    plt.show()

plot_metric(history,"accuracy")

================================================================================2020-04-21 08:17:27
step5: eval model......


Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_features (DenseFeature multiple                  64        
_________________________________________________________________
dense (Dense)                multiple                  3008      
_________________________________________________________________
dense_1 (Dense)              multiple                  4160      
_________________________________________________________________
dense_2 (Dense)              multiple                  65        
=================================================================
Total params: 7,297
Trainable params: 7,297
Non-trainable params: 0
_________________________________________________________________