padded shape not divisible by block shape

[Rosie-Brigham]

When I run either the eval or vis script I am given the error ensorflow.python.framework.errors_impl.InvalidArgumentError: padded_shape[0]=86 is not divisible by block_shape[0]=6
and tensorflow.python.framework.errors_impl.InvalidArgumentError: padded_shape[0]=86 is not divisible by block_shape[0]=6 respectively.

I am really new to all things TF and this has really stumped me - is it something that others have come across?

I have detailed the error in full on my repo - which I have modelled on this one.
Rosie-Brigham/fun-with-deeplab#1

Any help on this gratefully received!

Further info:

system information
What is the top-level directory of the model you are using:
DeepLab v3+

Have I written custom code (as opposed to using a stock example script provided in TensorFlow):
Yes, I have written some code to train a pretrained model with new images

OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
Mac OS Catalina

TensorFlow installed from (source or binary):
With pip3 binary installation

TensorFlow version (use command below):
1.15.0 CPU only

CUDA/cuDNN version:
Not used

GPU model and memory:
Not used

[heaversm]

I believe this is related to the size of the images you are processing. Do you know what size images you are using in eval / vis?

[dmeirap]

I'm having the same issue, did you solve it?

[heaversm]

@dmeirap - Did you set your --train_crop_size=1000,667 to the dimension of your actual images, and did you make sure all of the images you are training on match those dimensions?