GitHub · Where software is built

This repository was archived by the owner on Jul 7, 2023. It is now read-only.

Labels Milestones

KeyError when trying to use transformer_moe for language modelling

feature request

#830

· maremun opened

on May 23, 2018

feature request support for customize metric

feature request

#822

· xiang-deng opened

on May 22, 2018

Propose adding data generators for Allen Brain Atlas image modeling, SR, and in-painting problems

feature request

#820

· cwbeitel opened

on May 21, 2018

Is there any plan to make fast_decode available for dot_product_relative?

feature request

#778

· annisamansa opened

on May 9, 2018

help Passing hparams via config file

feature request

#772

· Hazepusik opened

on May 8, 2018

feature request Suggestions about super_lm

feature request

#758

· AranKomat opened

on Apr 30, 2018

Does the current existing code support non-teacherforcing training?

feature request

#749

· Sanqiang opened

on Apr 26, 2018

help Do not shuffle dataset

feature request

#741

· Danysolism opened

on Apr 24, 2018

Adding pre-trained models. Maybe to tensorflow hub also.

feature request

#682

· gaceladri opened

on Mar 31, 2018

Requesting Sparse Input Support - MOE layer

feature request

#665

· pgyrya opened

on Mar 23, 2018

Feature suggestion: two rand seeds for weights initialization and training data position

feature request

#629

· martinpopel opened

on Feb 28, 2018

Interactive decoding or t2t serving returns <EOS> and <pad> tags in the result.

feature request

#622

· ndvbd opened

on Feb 27, 2018