从零基础入门Tensorflow2.0 ----四、15.tf.data读取csv文件并与tf.keras结合使用

every blog every motto: Love is carefully designed lie.

0.前言

数据 -> csv文件 -> tf.data读取csv文件
实战tf.data的使用

1. 代码部分

1. 导入模块

import matplotlib as mpl
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import sklearn
import pandas as pd
import os
import sys
import time
import tensorflow as tf
from tensorflow import keras

print(tf.__version__)
print(sys.version_info)
for module in mpl,np,pd,sklearn,tf,keras:
    print(module.__name__,module.__version__)

2. 读取数据

from sklearn.datasets import fetch_california_housing

# 房价预测
housing = fetch_california_housing()

3. 划分样本

# 划分样本
from sklearn.model_selection import train_test_split

x_train_all,x_test,y_train_all,y_test = train_test_split(housing.data,housing.target,random_state=7)
x_train,x_valid,y_train,y_valid = train_test_split(x_train_all,y_train_all,random_state=11)

print(x_train.shape,y_train.shape)
print(x_valid.shape,y_valid.shape)
print(x_test.shape,y_test.shape)

4. 数据归一化

# 归一化
from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
x_train_scaled = scaler.fit_transform(x_train)
x_valid_scaled = scaler.transform(x_valid)
x_test_scaled = scaler.transform(x_test)

5. 产生csv文件

output_dir = 'generate_csv'
if not os.path.exists(output_dir):
    os.mkdir(output_dir)
    
def save_to_csv(output_dir,data,name_prefix,header=None,n_parts=10):
    
    path_format = os.path.join(output_dir,'{}_{:02d}.csv')
    filenames = []
    
    for file_idx,row_indices in enumerate(np.array_split(np.arange(len(data)),n_parts)):
        part_csv = path_format.format(name_prefix,file_idx)
        filenames.append(part_csv)
        
        with open(part_csv,'wt',encoding='utf-8') as f:
            if header is not None:
                f.write(header + '\n')
            for row_index in row_indices:
                f.write(','.join([repr(col) for col in data[row_index]]))
                f.write('\n')
    
    return filenames
    

train_data = np.c_[x_train_scaled,y_train]
valid_data = np.c_[x_valid_scaled,y_valid]
test_data = np.c_[x_test_scaled,y_test]

header_cols = housing.feature_names + ['MiddianHouseValue']

header_str = ','.join(header_cols)

train_filenames = save_to_csv(output_dir,train_data,'train',header_str,n_parts=20)
valid_filenames = save_to_csv(output_dir,valid_data,'valid',header_str,n_parts=10)
test_filenames = save_to_csv(output_dir,test_data,'test',header_str,n_parts=10)

6. 读取文件

6.1 打印文件名

# 读取上面生成的文件
# 打印文件名
import pprint
print("train filenames: ")
pprint.pprint(train_filenames)
print("valid filenames: ")
pprint.pprint(valid_filenames)
print("test filenames: ")
pprint.pprint(test_filenames)

6.2 读取文件

# 读取文件
# 1. filename -> dataset
# 2. read file -> dataset -> datasets -> megre
# 3. parse csv

filename_dataset = tf.data.Dataset.list_files(train_filenames)
for filename  in filename_dataset:
    print(filename)

n_readers = 5 # 并行度
dataset = filename_dataset.interleave(
    lambda filename : tf.data.TextLineDataset(filename).skip(1), # skip(1) 省略一行
    cycle_length = n_readers
)

for line in dataset.take(15):
    print(line.numpy())

7. 解析csv文件

# parse csv 字符串变成tensor
# tf.io.decode_csv(str,record_defaults,)

sample_str = '1,2,3,4,5'
record_defaults = [tf.constant(0,dtype=tf.int32)] * 5
parsed_fields = tf.io.decode_csv(sample_str,record_defaults)
print(parsed_fields)

sample_str = '1,2,3,4,5'
record_defaults = [tf.constant(0,dtype=tf.int32),
                  0,
                  np.nan,
                  'hello',
                  tf.constant([])]
parsed_fields = tf.io.decode_csv(sample_str,record_defaults)
print(parsed_fields)

7.2 测试

try:
    parsed_fields = tf.io.decode_csv(',,,,',record_defaults)
except tf.errors.InvalidArgumentError as ex:
    print(ex)

try:
    parsed_fields = tf.io.decode_csv('1,2,3,4,5,6,7',record_defaults)
except tf.errors.InvalidArgumentError as ex:
    print(ex)

8. 解析dataset中的一行

# 解析dataset 中的一行

def parse_csv_line(line,n_fields=9):
    defs = [tf.constant(np.nan)] * n_fields
    parsed_fields = tf.io.decode_csv(line,record_defaults=defs)
    x = tf.stack(parsed_fields[0:-1])
    y = tf.stack(parsed_fields[-1:])
    
    return x,y
    
parse_csv_line(b'-0.060214068004363165,0.7527628439249472,0.0835940301935345,-0.06250122441959183,-0.03497131082291674,-0.026442380178345683,1.0712234607868782,-1.3707331756959855,1.651',
              n_fields=9)

9. 整体

整体

1. filename -> dataset
2. read file -> dataset -> datasets -> megre
3. parse csv

# 整体
# 1. filename -> dataset
# 2. read file -> dataset -> datasets -> megre
# 3. parse csv

def csv_reader_dataset(filenames,n_readers=5,batch_size=32,n_parse_threads=5,
                      shuffle_buffer_size=10000):
    
    # 1. filename -> dataset
    dataset = tf.data.Dataset.list_files(filenames)
    dataset = dataset.repeat() # 重复多少次
    # 文件名转换成文本内容
    dataset = dataset.interleave(
            lambda filename:tf.data.TextLineDataset(filename).skip(1),
            cycle_length = n_readers
    )
    dataset.shuffle(shuffle_buffer_size)
    # 解析
    dataset = dataset.map(parse_csv_line,
                         num_parallel_calls=n_parse_threads)
    
    dataset = dataset.batch(batch_size)
    return dataset

# 测试
train_set = csv_reader_dataset(train_filenames,batch_size=3)
for x_batch,y_batch in train_set.take(2):
    print("x:")
    pprint.pprint(x_batch)
    print("y:")
    pprint.pprint(y_batch)
    

10. 搭建模型

# 搭建模型
model = keras.models.Sequential([
    keras.layers.Dense(30,activation='relu',input_shape=[8]),
    keras.layers.Dense(1),
    
])

# 打印model信息
model.summary()
# 编译
model.compile(loss='mean_squared_error',optimizer="sgd")
# 回调函数
callbacks = [keras.callbacks.EarlyStopping(patience=5,min_delta=1e-3)]

11. 训练

#训练
history = model.fit(train_set,validation_data=valid_set,
                    steps_per_epoch=11160 // batch_size,
                    validation_steps = 3870 // batch_size,
                    epochs=100,callbacks=callbacks)

12. 测试集上

model.evaluate(test_set,steps = 5160 // batch_size)