failure with custom model

[iandow]

I'm getting the following exception when I try to run the fx app with a custom model and label txt file:

2017-09-14 22:36:04.863769: W tensorflow/core/framework/op_kernel.cc:1152] Invalid argument: computed output size would be negative

I think that's a problem with tensorflow's Java API, but just thought I'd check to see if anyone else has seen it here.

[jesuino]

Hello Ian,

The code will need to be adjusted in order to support a new image! You need to adjust the input tensor, it was made for a specific model only. If you were able to run with your customer would you please share the code with us and how you did it?

Thanks.