How can I solve the input shape issue in CNN?

Question

I am using to classify the handwritten characters and following my CNN model. I have converted the images to gray scale and binarized. How can I solve this problem?

model = models.Sequential()

model.add(layers.Conv2D(filters=10,kernel_size=(3,3),padding = 'same', input_shape=(100, 100, 3),activation = 'relu'))
model.add(layers.Conv2D(30,(5,5),activation='relu'))

model.add(layers.MaxPooling2D(pool_size=(2,2)))

model.add(layers.Dropout(0.5))

model.add(layers.Conv2D(40,(3,3),activation='relu'))
model.add(layers.Conv2D(60,(3,3),activation='relu'))

model.add(layers.MaxPooling2D(pool_size=(2,2)))

model.add(layers.Flatten())

model.add(layers.Dense(180,activation='relu'))

model.add(layers.Dropout(0.5))

model.add(layers.Dense(100,activation='relu'))
model.add(layers.Dense(12,activation='softmax'))

Error

ValueError: Input 0 of layer sequential_41 is incompatible with the layer: expected ndim=4, found ndim=3. Full shape received: [None, 100, 100]

Answer 1

In your model you've stated the input shape to be (100,100,3) . The model is expecting a 3 channel image but you've said your input data is grayscale, meaning that it only has one channel. This means that your input data is of the shape (100,100,1) , which then is just (100,100) . With the addition of the batch_size dimension it becomes [None,100,100] which is the error you are getting.

Changing the input shape to (100,100,1) should do the trick.