Is it possible to predict a batch of images with coremltools mlmodel?

Question

Is it possible to predict a batch in mlmodel? If yes, how?

I convert a keras model to mlmodel, as presented in the documentation:

import coremltools as ct 
image_input = ct.ImageType(name='input', shape=(1, 224, 224, 3))
model = ct.convert(keras_model, inputs=[image_input])

Next, I load an image, resize it to (224, 224), transform it to PIL, and use it in prediction:

img_resized = cv2.resize(img_read, (224, 224))
PIL_image = Image.fromarray(np.uint8(img_resized))
img_heatmap = self.model.predict({'input': PIL_image})

Is it possible to do it a batch of images instead of single image?

I've tried to define a batch of input type in the conversion:

BATCH_SIZE = 8
image_input = ct.ImageType(name='input', shape=(BATCH_SIZE, 224, 224, 3))
model = ct.convert(keras_model, inputs=[image_input])

But it doesn't seem to change the model's input.

1. When I try to provide it with a numpy array of the BATCH_SIZE images ((8, 224, 224, 3)), I receive:

    return self.__proxy__.predict(data, useCPUOnly)
RuntimeError: value type not convertible

2. When I try to provide it with a list of BATCH_SIZE PIL images I receive:

RuntimeError: {
    NSLocalizedDescription = "Required input feature not passed to neural network.";
}

3. When I predict with a single image, the output is the same as the model with the BATCH_SIZE=1.

Is it possible to predict a batch? Thanks