Skip to content

Tensorflow can't handle my shape forms  #4

@Alkapo

Description

@Alkapo

So I tried on many env to run PoolNet project with the same data used to have the above results.
I have all the files (filtered.geojson etc... plus the image from GDBX) so usually I am using the same polygons.

When I come to the point to start my VG16 without any data I have some warnings but ...still it works.
After that point ! Everything crash with this particular message error that you can see in the first file.
In the beginning, I thought the main reason is the Keras lib which has changed or even Tenserflow.
That being said...When I tried to downgrade my Keras version Tensorflow will not follow especially with python3.5.

ResultFile.txt

Anyway, with all the changes made.... I still have the same problem.

If the code had made it with good results.....Obviously, the problem is on my side so I made a file with all my lib in my env with the version of each one of it.

LibEnvLocal.txt

Can anyone help me with that, please ?!
Thank you

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions