More than one prediction in multi-classification in Keras?

Question

I am learning about designing Convolutional Neural Networks using Keras. I have developed a simple model using VGG16 as the base. I have about 6 classes of images in the dataset. Here are the code and description of my model.

model = models.Sequential()
conv_base = VGG16(weights='imagenet' ,include_top=False, input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3))
conv_base.trainable = False
model.add(conv_base)
model.add(layers.Flatten())
model.add(layers.Dense(256, activation='relu', kernel_regularizer=regularizers.l2(0.001)))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(6, activation='sigmoid'))

Here is the code for compiling and fitting the model:

model.compile(loss='categorical_crossentropy',
        optimizer=optimizers.RMSprop(lr=1e-4),
         metrics=['acc'])
model.summary()

callbacks = [
    EarlyStopping(monitor='acc', patience=1, mode='auto'),
    ModelCheckpoint(monitor='val_loss', save_best_only=True, filepath=model_file_path)
]

history = model.fit_generator(
    train_generator,
    steps_per_epoch=10,
    epochs=EPOCHS,
    validation_data=validation_generator,
    callbacks = callbacks,
    validation_steps=10)

Here is the code for prediction of a new image

img = image.load_img(img_path, target_size=(IMAGE_SIZE, IMAGE_SIZE))
plt.figure(index)
imgplot = plt.imshow(img)

x = image.img_to_array(img)
x = x.reshape((1,) + x.shape)
prediction = model.predict(x)[0]
# print(prediction)

Often model.predict() method predicts more than one class.

[0 1 1 0 0 0]

I have a couple of questions

1. Is it normal for a multiclass classification model to predict more than one output?
2. How is accuracy measured during training time if more than one class was predicted?
3. How can I modify the neural network so that only one class is predicted?

Any help is appreciated. Thank you so much!

Division

Answer 1

You are not doing multi-class classification, but multi-label. This is caused by the use of a sigmoid activation at the output layer. To do multi-class classification properly, use a softmax activation at the output, which will produce a probability distribution over classes. Taking the class with the biggest probability (argmax) will produce a single class prediction, as expected.

Reneerenegade

Thank you for the explanation. That was it!. Can you explain how accuracy is measured by Keras train function? For example, during training, suppose the model predicted [0, 0.2, 0.4, 0.7, 0.1, 0]. Does keras look at class 3, with the accuracy of 0.7 and consider that as the output of the model to compare with the truth? – Division

This approach is not library specific. The general multi-class classification probability is to use softmax activation with n output classes, taking the "pick" to be the one of the highest probability. So in your case, yes class 3 is considered to be the selected class. Accuracy on a single sample is binary and averaged over your input. @TMS. – Viva