如何在Keras中将'mnist.load_data()'返回值的格式更改为'mnist_train.csv'? [英] how to change the format of the return value of 'mnist.load_data()' to 'mnist_train.csv' in Keras?

问题描述

我使用Keras.

(X_train, y_train), (X_test, y_test) = mnist.load_data()

X_train'的形状为(number_of_training_sample,224,224,3)

X_train' shape is (number_of_training_sample,224,224,3)

Y_train的形状为(number_of_training_sample, 10)

Y_train's shape is (number_of_training_sample, 10)

功能和标签在不同的ndarray中分开. 但是我想将这些ndarrays更改为'mnist_train.csv'格式. ( https://raw.githubusercontent.com/sjwhitworth/golearn /master/examples/datasets/mnist_train.csv )

Features and labels are separated in different ndarray. BUT I want to change these ndarrays to 'mnist_train.csv' format. (https://raw.githubusercontent.com/sjwhitworth/golearn/master/examples/datasets/mnist_train.csv)

我自己的数据集的格式与返回值'mnist.load_data()'完全相同.但是在我要使用的代码中只能使用"mnist_train.csv"格式.

My own dataset is totally same format with the return value of 'mnist.load_data()'. BUT only 'mnist_train.csv' format is worked in the code that I want to use.

您能让我知道如何更改格式吗???? (很抱歉，这个问题很简单.我是numpy和python的新手.)

Could you let me know how to change the format???? (Sorry for really simple question.. I am a newbie of numpy and python.)

推荐答案

(X_train, y_train), (X_test, y_test) = mnist.load_data()

将数据作为numpy数组加载

loads data as numpy arrays

'mnist_train.csv'是硬盘上的.csv文件存储，我们通常使用pandas库读取

'mnist_train.csv' is a .csv file store on a hard disc which we usually read with pandas library

import pandas as pd
X_train = pd.read_csv('filename.csv')

Pandas读取一个.csv作为数据框(如果有列名称，则带有列名称) 使用keras，您需要numpy数组，因此

Pandas reads a .csv as a dataframe (with column names if there are such) Using keras you need numpy arrays, so the

(X_train, y_train), (X_test, y_test) = mnist.load_data()

应该可以正常工作

发布您正在使用的代码以及所产生的错误以供进一步解释

Post the code you are using and the errors that is produced for further explaination

这篇关于如何在Keras中将'mnist.load_data()'返回值的格式更改为'mnist_train.csv'?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！