search

IDSIA / brainstorm

Fast, flexible and fun neural networks.
Other
1.3k stars 152 forks source link

Addressing outputs in hooks and such #59

Closed Qwlouse closed 8 years ago

Qwlouse commented 8 years ago

In hooks and for net.get_output you currently need to specify Layer.outname. But in many other places we use the full path Layer.outputs.outname. These two different ways of addressing outputs can be confusing. So should we unify this to the long version? It'd be more clear but also more verbose.

flukeskywalker commented 8 years ago

Explicit is better than implicit. However, get_output should only allow you to get the outputs :P