如果我尝试拟合来自pandas数据框的数据，如何将input_shape赋予卷积神经网络？

Question

我有一个733999个样本和5个功能的Pandas数据X_train X_train。

model = Squential()    
model.add(Conv2D(filters = 32, kernel_size = (5,5),padding = 'Same', 
             activation ='relu', input_shape = (?,?)))

这是我遇到麻烦的第一层。 所有教程都使用了图像，它们只是将高度，宽度和通道作为input_shape的参数传递。 在熊猫数据框的情况下，我很难给出输入形状。 任何帮助都非常感谢。

Answer 1

这是有关如何对数据使用CNN的示例

对于您拥有的数据，我仍然不建议使用这种类型的网络

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Reshape
import pandas as pd
import numpy as np

## Dummy data
data = {'0': [1, 2, 3], '1': [3, 4, 3], '2':[0,1, 3], '3':[0,1,3], '4':[0,1,3], '5':[0,1,3]}
X_train = pd.DataFrame(data=data)

model = Sequential()
model.add(Reshape((1,X_train.shape[1],1)))
model.add(Conv2D(filters = 32, kernel_size = (1,5),padding = 'Same',
             activation ='relu', input_shape = (1,X_train.shape[1],1)))

model.add(MaxPooling2D(pool_size = (1,6), strides=(1,2)))

model.add(Flatten())

model.add(Dense (500, activation='relu'))
model.add(Dense (1, activation='relu'))
model.compile(loss='binary_crossentropy', optimizer='adam',
              metrics=['accuracy'])

## Training and testing with dummy data just to prove that it's working
model.fit(np.array(X_train), np.array([0,1,1]), nb_epoch=4, validation_data=(np.array(X_train), np.array([0,1,1])))