model.predict_on_batch(training_data['inputs']).round(1) gives error AttributeError: 'tensorflow.python.framework.ops.EagerTensor' object has no attribute 'round' I have tried restarting the kernel and also updated TensorFlow to v2.1.0. I also have tried adding eager execution.But,as the method is automatic in TF2,it throws up another error.

Hi Manav, Thanks for reaching out to us. In order to use round on tensors of TensorFlow 2.1, we need to get their numpy arrays: model.predict_on_batch(training_data['inputs']).numpy().round(1) Best,The 365 Team

Follow this topic

Manav Patel

Last answered:

23 Apr 2020

Posted on:

12 Apr 2020

Error faced while extracting outputs

1 answers ( 0 marked as helpful)

365 Team

Posted on:

23 Apr 2020

Error faced while extracting outputs

Submit an answer