`batch_size` > 1 requires an `image_resizer` of `fixed_shape_resizer` in Tensorflow

[dmitrii_fediuk]

Tensorflow requires each input array to have the same dimensionality, which means that any batch_size > 1 requires an image_resizer of fixed_shape_resizer

blog.algorithmia.com/deep-dive-into-object-detection-with-open-images-using-tensorflow