我面临分类分类器
模型输入形状的问题 x y
[1,2,3] [0]
[2,3,5] [1]
[2,1,6] [2]
[1,2,3] [0]
[2,3,5] [0]
[2,1,6] [2]
然后我将y标签更改为分类为
y
[1,0,0]
[0,1,0]
[0,0,1]
[1,0,0]
[1,0,0]
[0,0,1]
我的x_train形状是(6000,3) y_train形状是(6000,3) x_test形状是(2000,3) y_test形状是(2000,3)
我尝试了这个模型并获得了价值错误
model=sequential()
model.add(Dense(1, input_shape(3,), activation="softmax"))
model.compile(Adam(lr=0.5), 'categorical_crossentropy', metrics=['accuracy'])
model.fit(X_train,y_train,epochs=50, verbose=1)
Value error: Error when checking target: expected dense_1 to have shape(None,1) but got array with shape (6000,3)
我不明白这个错误。帮我理清这个
答案 0 :(得分:0)
您的网络需要一个与输出类数相匹配的输出层。你可以这样做
X_train = np.zeros((10,3))
y_train = np.zeros((10,))
X_test = np.zeros((10,3))
y_test = np.zeros((10,))
num_classes = 3
y_train_binary = keras.utils.to_categorical(y_train, num_classes)
y_test_binary = keras.utils.to_categorical(y_test, num_classes)
input_shape = (3,)
model = Sequential()
model.add(Dense(16, activation='relu',input_shape=input_shape))
model.add(Dense(num_classes, activation='softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='rmsprop',
metrics=['mae'])
model.summary()
history=model.fit(X_train,
y_train_binary,
epochs=5,
batch_size=8,
validation_data=(X_test, y_test_binary))