无法在 keras DNN 上正确上传 csv 数据集

Question

我将 [Kaggle 数据集][1] 用于 mnist 手语。总共有 785 列，包括带有 CSV 数据集标签的一列。将 CSV 用于图像而不是真实图像也是一个好主意

以下代码运行良好，直到 mode.fit() 出错

"""CSV_MODEL.ipynb

Automatically generated by Collaboratory.

The original file is located at
    https://colab.research.google.com/drive/1u8GDJe-sWtz12YO7YusClJR9UeDJ852Y
"""

from google.colab import files
uploaded = files.upload()

from keras.models import Sequential
from keras.layers import Dense
import numpy

numpy.random.seed(23)

import csv
import numpy
filename = 'sign_mnist_train.csv'
raw_data = open(filename, 'rt')
reader = csv.reader(raw_data, delimiter=',', quoting=csv.QUOTE_NONE)
x = list(reader)
data = numpy.array(x).astype('float')
print(data.shape)

print(data.shape[1])
X = data[:,0:784]
Y = data[:,784]

print(X.shape[1])
print(X)

model = Sequential()
model.add(Dense(512,input_dim = 784,activation='relu'))
model.add(Dense(24,activation='softmax'))
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
print(model.summary())

model.fit(X,Y,epochs=100,batch_size=20)```

它给出了这样的错误

ValueError: Error when checking target: expected dense_34 to have shape (24,) but got array with shape (1,)
[1]: https://www.kaggle.com/datamunge/sign-language-mnist

Answer 1

当你使用时

loss='categorical_crossentropy'

你必须先对标签进行编码（一个热编码）

keras.utils.to_categorical(Y)

或者您可以将损失函数更改为

loss='sparse_categorical_crossentropy'