错误 - 使用Keras和我自己的csv数据训练神经网络

Question

美好的一天！我想使用Keras python库训练神经网络我想制作4个输入神经元和1个输出神经元。我想用我自己的csv文件和数字：这是它

my_5_input_numbers.csv
0.3,0.5,0.6,0.7,1
0.4,0.6,0.7,0.8,0
0.5,0.7,0.8,0.9,1

我使用numpy来读取csv并制作列车矩阵。这是代码和错误

import numpy as np
np.set_printoptions(threshold=np.inf)
from keras.datasets import boston_housing
from keras.models import Sequential
from keras.layers import Dense
data_common=np.genfromtxt('my_5_input_numbers.csv',delimiter=',')
"""
data_common=array([[ 0.3,  0.5,  0.6,  0.7,  1. ],
       [ 0.4,  0.6,  0.7,  0.8,  0. ],
       [ 0.5,  0.7,  0.8,  0.9,  1. ]])
data_common.shape=(3,5)       
"""

X_train=data_common[:,-1]#X_train.shape=(3,4)
y_train=data_common[0:4,-1]#y_train.shape=(3,)
y_train=y_train.reshape(3,1)
model = Sequential()
model.add(Dense(128,input_dim=4,activation='relu'))
model.add(Dense(1,activation="softmax"))
model.compile(optimizer='adam', loss='mse', metrics=['mae'])
# train nn
model.fit(X_train, y_train, batch_size=200, epochs=25, validation_split=0.2, verbose=2)
#<---Error:File "D:\NetbeansPythonProjects\testDiffrentCode\src\testKeras.py", line 16, in <module>
#    model.fit(X_train, y_train, batch_size=200, epochs=25, validation_split=0.2, verbose=2)
#ValueError: 
#Error when checking input:
#expected dense_1_input to have shape (None, 4) but got array with shape (3, 1)

Answer 1

你只是搞砸了你的输入：

X_train=data_common[:,-1] # <--- X_train.shape was actually (3,) not (3,4).
y_train=data_common[0:4,-1] # <--- This was wrong as well.
y_train=y_train.reshape(3,1)

本来应该的

X_train=data_common[:,0:4]
y_train=data_common[:,-1] 
y_train=y_train.reshape(3,1)

numpy的索引是第一行，然后是列。