运行 Keras 神经网络时出现结果形状问题

Question

I want to make prediction with Keras Neural Network. My output data has 3 different values -1,0,1. When I run my NN I get the error:

ValueError: Error when checking target: expected dense_35 to have shape (3,) but got array with shape (1,)

Then I tried to do:

from tensorflow.python.keras.utils import to_categorical
results = to_categorical(results)

But again I get the same error:

ValueError: Error when checking target: expected dense_35 to have shape (3,) but got array with shape (2,)

What am I doing wrong? This is my code:

features = df.iloc[:,-8:]
results = df.iloc[:,-9]

x_train, x_test, y_train, y_test = train_test_split(features, results, test_size=0.3, random_state=42)


model = Sequential()
model.add(Dense(64, input_dim = x_train.shape[1], activation = 'relu')) # input layer requires input_dim param
model.add(Dense(32, activation = 'relu'))
model.add(Dense(16, activation = 'relu'))
model.add(Dense(3, activation = 'softmax'))

model.compile(loss="categorical_crossentropy", optimizer= "adam", metrics=['accuracy'])

# call the function to fit to the data training the network)
es = EarlyStopping(monitor='val_loss', min_delta=0.001, patience=0, verbose=1, mode='auto')
model.fit(x_train, y_train, epochs = 10, shuffle = True, batch_size=128, validation_data=(x_test, y_test), verbose=2, callbacks=[es])

Answer 1

results = df.iloc[:,-9] you're choosing 1-d output (shape: (rows,1)), but your last layer has 3 units model.add(Dense(3, activation = 'softmax')) .

So, your result must have shape: (rows, 3) not (rows, 1).

I see your result has values -1, 0, 1. Just add one so that they are 0, 1, 2. That's why you're getting error with to_categorical ; according to the docs , it expects

y : class vector to be converted into a matrix (integers from 0 to num_classes).

So go for

results = results + 1

Then, apply to_categorical .

After that fit should work fine.