Organizing transfer learning
Basically I organized the keyword arguments for the input networks used in the estimator.
Now, we can expect that the network has this function signature:
inputs, reuse=False, is_training_mode = True, trainable_variables=True
where,
- reuse: If you want to reuse the tensorflow variables
- is_training_mode: If you are in training mode or not.
- trainable_variables: IF the set of variables in the network function are trainable
Tests are passing.
Edited by Tiago de Freitas Pereira
Merge request reports
Activity
added 1 commit
- 22569cc1 - Organized the transfer learning variables for the estimators
assigned to @amohammadi
@tiago.pereira usually architecture functions accept an input called
mode
: https://github.com/tensorflow/models/blob/edcd29f2dbb4b3eaed387fe17cb5270f867aec42/official/mnist/mnist.py#L99 instead of a flagis_training_mode
. Don't you think staying close to tensorflow official signature would be better?added 1 commit
- 3770d596 - Renamed the keyword argumente is_training_mode to mode and use the…
Just switched the variables @amohammadi> I will work on your merge now.
mentioned in commit 5fb18ca1
Please register or sign in to reply