Incompatible inputs of layers (ndim=4, found ndim=3)

Question

I tried to recreate this Flower Recognition CNN in Keras. The model seems to work, at least in the notebook (while getting predictions on the validation set), but I need to use the model somewhere else. Photos are 150x150 and this is how I build the CNN:

model = Sequential()
model.add(Conv2D(filters=32, kernel_size=(5,5), padding='Same', activation='relu', input_shape=(150, 150, 3)))
model.add(MaxPooling2D(pool_size=(2,2)))

model.add(Conv2D(filters=64, kernel_size=(3,3), padding='Same', activation='relu'))
model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2)))

model.add(Conv2D(filters=96, kernel_size=(3,3), padding='Same', activation='relu'))
model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2)))

model.add(Conv2D(filters=96, kernel_size=(3,3), padding='Same', activation='relu'))
model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2)))

model.add(Flatten())
model.add(Dense(512))
model.add(Activation('relu'))
model.add(Dense(5, activation="softmax"))

Still, when I try to predict the result locally with sample photos and model loaded :

model = model_from_json(open("model.json", "r").read())
model.load_weights('model.h5')

img = cv2.imread('/path/to/the/dir/testimage.jpg')
img = cv2.resize(img, (150, 150))
data = np.array(img)
model.summary()
result = model.predict(data)

I receive this error:

ValueError: Input 0 of layer sequential_1 is incompatible with the layer: expected ndim=4, found ndim=3. Full shape received: [None, 150, 3]

The output of model summary is:

Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_1 (Conv2D)            (None, 150, 150, 32)      2432      
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 75, 75, 32)        0         
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 75, 75, 64)        18496     
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 37, 37, 64)        0         
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 37, 37, 96)        55392     
_________________________________________________________________
max_pooling2d_3 (MaxPooling2 (None, 18, 18, 96)        0         
_________________________________________________________________
conv2d_4 (Conv2D)            (None, 18, 18, 96)        83040     
_________________________________________________________________
max_pooling2d_4 (MaxPooling2 (None, 9, 9, 96)          0         
_________________________________________________________________
flatten_1 (Flatten)          (None, 7776)              0         
_________________________________________________________________
dense_1 (Dense)              (None, 512)               3981824   
_________________________________________________________________
activation_1 (Activation)    (None, 512)               0         
_________________________________________________________________
dense_2 (Dense)              (None, 5)                 2565      
=================================================================
Total params: 4,143,749
Trainable params: 4,143,749
Non-trainable params: 0

I cannot understand why would it receive a different shape. I have read that maybe Reshape would do the thing, but I am using fit_generator to fit on training set and I am not sure if that's approachable.

Answer 1

Make sure you have the first dimension as 1 (1 image to predict). Your model is expecting the None dimension to be number of samples, 2nd + 3rd to be the image resolution, and 4th to be your RGB channels for the image.

data = np.expand_dims(data, axis=0)

Will add an extra dimension to your first axis.

See: How can I add new dimensions to a Numpy array?