Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Oct 12, 2021

• Updated/Added CODEOWNERS with ECCN

  svc-scm

  committed
  043ab56

  Copy full SHA for 043ab56

Commits on Jan 23, 2019

• Add links for single-task pretrained models

  bmccann

  authored
  1e9605f

  Copy full SHA for 1e9605f

Commits on Jan 10, 2019

• prepend root before save

  bmccann

  committed
  cd997f2

  Copy full SHA for cd997f2

Commits on Jan 9, 2019

• log_dir needs root prefix

  bmccann

  committed
  eed545b

  Copy full SHA for eed545b

• options for sgd

  bmccann

  committed
  81b7ea7

  Copy full SHA for 81b7ea7

• unique val tasks

  bmccann

  committed
  27bb192

  Copy full SHA for 27bb192

Commits on Jan 8, 2019

• separate out root dir; add option for experiment name

  bmccann

  committed
  287b911

  Copy full SHA for 287b911

Commits on Dec 12, 2018

• Update README.md

  bmccann

  authored
  07dd886

  Copy full SHA for 07dd886

• bugs in overwrite; new best mqan model

  bmccann

  committed
  c096a1f

  Copy full SHA for c096a1f

Commits on Dec 10, 2018

• moving schema raw files to s3; updated pretrained models

  bmccann

  committed
  1dc5f7d

  Copy full SHA for 1dc5f7d

• back compat predict.py

  bmccann

  committed
  1bd922b

  Copy full SHA for 1bd922b

• more dynamic elmo use cases

  bmccann

  committed
  9a4493b

  Copy full SHA for 9a4493b

• elmo works better w/o dropout

  bmccann

  committed
  37003d2

  Copy full SHA for 37003d2

• adding elmo naively as a replacement

  bmccann

  committed
  6260c9f

  Copy full SHA for 6260c9f

• better to have tensorboard logging on as default

  bmccann

  committed
  3aecae0

  Copy full SHA for 3aecae0

Commits on Nov 28, 2018

• Remove CUDA from the command to build docker image

  

  daminisatya

  authored and

  bmccann

  committed
  6b31d8f

  Copy full SHA for 6b31d8f

Commits on Nov 17, 2018

• Update README.md

  bmccann

  authored
  39720d0

  Copy full SHA for 39720d0

• Update README.md

  bmccann

  authored
  42f1bad

  Copy full SHA for 42f1bad

• CoVe results

  bmccann

  authored
  ae8d64b

  Copy full SHA for ae8d64b

• better predict; better MQAN+CoVe

  bmccann

  committed
  fe58004

  Copy full SHA for fe58004

Commits on Oct 26, 2018

• repairing docker commands in readme

  bmccann

  authored
  9447f41

  Copy full SHA for 9447f41

Commits on Oct 25, 2018

• Update README.md

  bmccann

  authored
  72aeb19

  Copy full SHA for 72aeb19

• Fix embedding layer implementation

  Show description for c0e4f70
  

  todpole3

  authored and

  bmccann

  committed
  c0e4f70

  Copy full SHA for c0e4f70

• getting rid of .data in models/common.py

  bmccann

  committed
  4523bca

  Copy full SHA for 4523bca

• ptr-gen mem efficient dist; only trainable params

  bmccann

  committed
  2c837ea

  Copy full SHA for 2c837ea

• pg self_attended_context in greedy

  bmccann

  committed
  754d76f

  Copy full SHA for 754d76f

• Reducing train_batch_tokens; 10k OOMs on 0.4.1

  bmccann

  committed
  f835558

  Copy full SHA for f835558

• mv save out of val; add arg for no tensorboard logging

  bmccann

  committed
  38ce8ac

  Copy full SHA for 38ce8ac

• val loss item

  bmccann

  committed
  80408b9

  Copy full SHA for 80408b9

• updating torch03 images with cove

  bmccann

  committed
  741fb6f

  Copy full SHA for 741fb6f

• updated README re: docker images for devices and pytorch versions

  bmccann

  committed
  c5719e7

  Copy full SHA for c5719e7

• update devices

  bmccann

  committed
  2bcd810

  Copy full SHA for 2bcd810

• updating dockerfiles

  bmccann

  committed
  b120715

  Copy full SHA for b120715

• touching up predict

  bmccann

  committed
  eab80b6

  Copy full SHA for eab80b6

• mv to cpu for save; multinli.in.out regression

  bmccann

  committed
  fb0dcaa

  Copy full SHA for fb0dcaa

Pagination

Previous

Next