You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I was running the adanet_objective notebook and my result shows that the architecture is Architecture: b'| 1_layer_dnn |' instead of the expected output in the example notebook of " Architecture: b'| 1_layer_dnn |' '| 2_layer_dnn |' '| 3_layer_dnn |'.
I tried setting _measure_complexity to 0 to see what would happen and still got the output of Architecture: b'| 1_layer_dnn |', with the only single subnetwork.
Any input would be appreciated
The text was updated successfully, but these errors were encountered:
Hi, i see the same behavior on 0.8.0 version, but i see, that network is ok in architecture json files, tensorflow 1.15.
Update:
Everything ok on adanet 0.7.0
Hi,
I was running the adanet_objective notebook and my result shows that the architecture is Architecture: b'| 1_layer_dnn |' instead of the expected output in the example notebook of " Architecture: b'| 1_layer_dnn |' '| 2_layer_dnn |' '| 3_layer_dnn |'.
I tried setting _measure_complexity to 0 to see what would happen and still got the output of Architecture: b'| 1_layer_dnn |', with the only single subnetwork.
Any input would be appreciated
The text was updated successfully, but these errors were encountered: