What is "incompatibility of min_ndim and expected ndim" in 3D CNN(Convolutional NN)?

Question

I found many answers about similar questions, but they are almost about 'removing of flatten layer'. However I didn't use any flatten layer for input layer. So through searching I changed () into [] along the whole codes like this:

model.add(Conv3D(32, kernel_size = (3,3,3), input_shape = (32, 32,16), activation = 'relu'))

dat = np.load("/content/drive/MyDrive/trn_dat.npy")
dat = dat/255.0

lbl = np.load("/content/drive/MyDrive/trn_lbl.npy")
lbl = keras.utils.to_categorical(lbl, 2)

x_train, x_vd, y_train, y_vd = train_test_split(dat, lbl, test_size = 0.9)

# model creating

model = tf.keras.Sequential()
model.add(BatchNormalization(center = True, scale = True))
model.add(Conv3D(32, kernel_size = [3,3,3], input_shape = [32, 32,16], activation = 'relu'))
model.add(MaxPooling3D(pool_size = (2,2,2)))
model.add(Conv3D(32, kernel_size = [3,3,3], input_shape = [32, 32,16], activation = 'relu'))
model.add(MaxPooling3D(pool_size = (2,2,2)))
model.add(Dropout(0.2))

model.add(Conv3D(64, kernel_size = [3,3,3], input_shape = [32, 32,16], activation = 'relu'))
model.add(MaxPooling3D(pool_size = (2,2,2)))
model.add(Conv3D(64, kernel_size = [3,3,3], input_shape = [32, 32,16], activation = 'relu'))
model.add(MaxPooling3D(pool_size = (2,2,2)))
model.add(Dropout(0.25))

model.add(Conv3D(128, kernel_size = [3,3,3], input_shape = [32, 32,16], activation = 'relu'))
model.add(MaxPooling3D(pool_size = (2,2,2)))
model.add(Conv3D(128, kernel_size = [3,3,3], input_shape = [32, 32,16], activation = 'relu'))
model.add(MaxPooling3D(pool_size = (2,2,2)))
model.add(Dropout(0.25))

model.compile(loss = 'categorical_entropy', optimizer = 'adam', metrics = ['accuracy'])
early_stopping_callback = EarlyStopping(monitor = 'val_loss', patience =10)
path4 = model.fit(x_train, y_train, validation_data = (x_vd, y_vd), epochs = 100, callbacks = [early_stopping_callback])

And this is my error code:

/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
    971           except Exception as e:  # pylint:disable=broad-except
    972             if hasattr(e, "ag_error_metadata"):
--> 973               raise e.ag_error_metadata.to_exception(e)
    974             else:
    975               raise

ValueError: in user code:

    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:806 train_function  *
        return step_function(self, iterator)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:796 step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:1211 run
        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2585 call_for_each_replica
        return self._call_for_each_replica(fn, args, kwargs)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2945 _call_for_each_replica
        return fn(*args, **kwargs)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:789 run_step  **
        outputs = model.train_step(data)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:747 train_step
        y_pred = self(x, training=True)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/base_layer.py:985 __call__
        outputs = call_fn(inputs, *args, **kwargs)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/sequential.py:386 call
        outputs = layer(inputs, **kwargs)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/base_layer.py:976 __call__
        self.name)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/input_spec.py:196 assert_input_compatibility
        str(x.shape.as_list()))

    ValueError: Input 0 of layer conv3d_71 is incompatible with the layer: : expected min_ndim=5, found ndim=4. Full shape received: [None, 32, 32, 16]

Then what is the min_ndim? Is it one of hyperparameters that I can define?

Finally, how can I solve this error?

Answer 1

It means you need to feed 4D images (ndim = 5 because of the batch_size). You can reshape your dataset easily :

X_train=np.reshape(x_train,(x_train.shape[0], x_train.shape[1],
                            x_train.shape[2],x_train.shape[3],1))

and change input_shape (32,32,16) to (32,32,16,1).

Additionally, the input_shape is only valid for your first weight layer, you have to take off the others otherwise it will raise an error