Issues

• LossView.__init__ should call loss.losses before loss.updates

  bug
  SMART-Lab/smartlearner#72 · Filter by author ppoulin91 opened on Feb 23, 2016
  1

• Using multiple losses/datasets on the same model grows the update dictionary

  bug
  SMART-Lab/smartlearner#71 · Filter by author ppoulin91 opened on Feb 23, 2016

• When stopping criteria raise TrainingExit remaining tasks might be skipped

  SMART-Lab/smartlearner#61 · Filter by author MarcCote opened on Jan 15, 2016

• Experiments, Benchmark and Hyperoptimizer

  enhancement
  SMART-Lab/smartlearner#54 · Filter by author ASalvail opened on Sep 30, 2015
  1

• Make name parameter of sharedX mandatory

  SMART-Lab/smartlearner#52 · Filter by author MarcCote opened on Sep 21, 2015

• Add a meta trainer

  enhancement
  SMART-Lab/smartlearner#44 · Filter by author ASalvail opened on Aug 31, 2015

• Add docstring

  enhancement
  SMART-Lab/smartlearner#11 · Filter by author MarcCote opened on May 22, 2015
  13

• Define the interface for Model.py

  SMART-Lab/smartlearner#9 · Filter by author MarcCote opened on May 22, 2015
  1

• Re-think the notion of Task

  SMART-Lab/smartlearner#8 · Filter by author MarcCote opened on May 22, 2015

• Define the Optimizer.py

  enhancement
  SMART-Lab/smartlearner#7 · Filter by author MarcCote opened on May 22, 2015
  3

• Define the interface of Dataset.py

  SMART-Lab/smartlearner#4 · Filter by author MarcCote opened on May 21, 2015
  3