for-ai / CipherGAN

TensorFlow implementation of CipherGAN
https://openreview.net/forum?id=BkeqO7x0-
MIT License
121 stars 24 forks source link

How to specify prediction_key for Metric #5

Closed MaksymDel closed 5 years ago

MaksymDel commented 6 years ago

Hi!

When I run an example from the readme with Accuracy metric in addition to the default MSE, I get the following error:

ValueError: MetricSpec without specifiedValueError: MetricSpec without specified prediction_key requires predictions tensor or single element dict, got {'X': <tf.Tensor 'cycle_gan/transforms/G/Reshape_1:0' shape=(64, 100, 202) dtype=float32>, 'Y': <tf.Tensor 'cycle_gan/transforms/F/Reshape_1:0' shape=(64, 100, 202) dtype=float32>} requires predictions tensor or single element dict, got {'X': <tf.Tensor 'cycle_gan/transforms/G/Reshape_1:0' shape=(64, 100, 202) dtype=float32>, 'Y': <tf.Tensor 'cycle_gan/transforms/F/Reshape_1:0' shape=(64, 100, 202) dtype=float32>}

How does the correct prediction_key specification for the README example look like?

Best, Maksym

huangsicong commented 6 years ago

Thanks for the issue. We ran the code again and it was fine. What was the exact command you used?

Thanks, Sheldon

aidangomez commented 5 years ago

Closing due to inactivity. Please reopen if necessary.