Custom model compression

[mazzarellaf]

Hi,

I have successfully uploaded a custom model in .h5 format. Whenever I try to compress that model, I keep getting the same error stating:

"Error is occurred in server. If this error continues, please send us an email..."

Please, may I have some hints on how to fix that issues?

Many thanks!

[NetsPresso-admin]

We had a discussion by email after receiving mazzarellaf's account information.

Currently, the TransposeConv layer is not supported for TensorFlow models.
It is supported with PyTorch models and we are going to improve compatibility for TensorFlow models soon.

Sorry for the inconvenience and will let you know when your models become compatible!