How to solve "ValueError: Error when checking target: expected dense_20 to have shape (24,) but got array with shape (1,)"?

0
votes

I'm working on a sign language interpretation using deep learning & for that, i'm building a CNN and i am getting an error like,

ValueError: Error when checking target: expected dense_20 to have shape (24,) but got array with shape (1,)

structure of my neural net: 

Layer (type)                 Output Shape              Param #   

conv2d_62 (Conv2D)           (None, 64, 64, 64)        1088      
_________________________________________________________________
conv2d_63 (Conv2D)           (None, 32, 32, 64)        65600     
_________________________________________________________________
dropout_31 (Dropout)         (None, 32, 32, 64)        0         
_________________________________________________________________
conv2d_64 (Conv2D)           (None, 32, 32, 128)       131200    
_________________________________________________________________
conv2d_65 (Conv2D)           (None, 16, 16, 128)       262272    
_________________________________________________________________
dropout_32 (Dropout)         (None, 16, 16, 128)       0         
_________________________________________________________________
conv2d_66 (Conv2D)           (None, 16, 16, 256)       524544    
_________________________________________________________________
conv2d_67 (Conv2D)           (None, 8, 8, 256)         1048832   
_________________________________________________________________
flatten_11 (Flatten)         (None, 16384)             0         
_________________________________________________________________
dropout_33 (Dropout)         (None, 16384)             0         
_________________________________________________________________
dense_19 (Dense)             (None, 512)               8389120   
_________________________________________________________________
dense_20 (Dense)             (None, 24)                12312

code:

    model = Sequential()

    model.add(Conv2D(64, kernel_size=4, strides=1, activation='relu', input_shape = (64,64,1),padding = 'same'))
    model.add(Conv2D(64, kernel_size=4, strides=2, activation='relu',padding = 'same'))
    model.add(Dropout(0.2))

    model.add(Conv2D(128, kernel_size=4, strides=1, activation='relu',padding = 'same'))
    model.add(Conv2D(128, kernel_size=4, strides=2, activation='relu',padding = 'same'))
    model.add(Dropout(0.2))

    model.add(Conv2D(256, kernel_size=4, strides=1, activation='relu',padding = 'same'))
    model.add(Conv2D(256, kernel_size=4, strides=2, activation='relu',padding = 'same'))

    model.add(Flatten())
    model.add(Dropout(0.3))
    model.add(Dense(512, activation='relu'))
    model.add(Dense(24, activation='softmax'))

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

    print(model.summary())

    model.fit(x_train, y_train, validation_data=(x_test, y_test),batch_size=64,epochs=8)

dimensions of arrays used:
x_train: (3977, 64, 64, 1)
y_train: (3977, 1)
x_test: (995, 64, 64, 1)
y_test: (995, 1)

1
python-3.xkerasdeep-learningconv-neural-network
you are using model.add(Dense(24, activation='softmax')) at the end of your network so it expect your output is in shape of (24,) and you are feeding y_train with shape of (1,) If I am not wrong you are trying to do classification for 24 class? – Krunal V
@kruxx, yes i'm trying to classify 24 gestures. – tarang ranpara
Okay, try answer of Thibault Bacqueyrisses, it will work for you. – Krunal V
you need to use loss='sparse_categorical_crossentropy' instead of loss='categorical_crossentropy' . – Vishnuvardhan Janapati

1 Answers

1
votes

Your last layer output shape need to match your label's vector shape
So you need to one_hot encode your y_train in order to fit your network.

You can do this like that :

from keras.utils import to_categorical
y_train = to_categorical(y_train, 24)

This will encode each of your label to a vector of size 24 (or whatever you need), filled with 0s and a 1 at the position of the corresponding label.

To learn more about it :
https://keras.io/utils/