Dmitry Gordin

it neither works when I do not touch output node. (don't specify as image type, set flexible size)

@aseemw tested with MultiArray input with dynamic size – it work just perfect! (while keeping output as image) ```python size_l = 128 size_u = 2048 flexible_shape_utils.set_multiarray_ndshape_range(spec, INPUT_NODE, [1,3,size_l,size_l], [1,3,size_u,size_u]) ```...

@3DTOPO Thank you for workaround with Accelerate! How do you think is it the way CoreML convert UIImage internally? And could you provide small snippet how to use vImageConvert_ARGB8888toPlanarF in...

@alexrkr haven't got so far. Need to run the model at least somehow. So in your case model works on ANE when converted with fixed size but use CPU/GPU when...

@paulaksm @shariq-audiofocus have you tried structural pruning instead? if you worry about storage only why not to consider gzip compressing for the model file? (like cryptopp gzip)

Also please update Colab

@saddif I'm not sure about your case but I was able to overcome this error by putting the model through https://github.com/daquexian/onnx-simplifier Hope it helps

@Daniel-Ri you need CUDA installed on you PC, go to [developer.nvidia.com](https://developer.nvidia.com) and follow steps for your system, eg: 

Thank you for your ultra fast response! @zvezdochiot as long as we have very large datasets like [places2](http://places2.csail.mit.edu/) (10+ million photos) and [CelebA](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html) (200+ thousands of faces) we can expect...

Thank you for such fast answer! Initially I started investigation because I got pretty good results for overall face shape but poor results for eyes/iris region. So I though of...