Commits

Breadcrumbs

History for

ondevelop

User selector

All users

All time

Commit History

Commits on Feb 11, 2022

• 2022-02-11_10:52

  

  johnson

  committed
  54257fa

  Copy full SHA for 54257fa

Commits on Jan 29, 2022

• fix infer.py when use_faster_tokenizer == False (#1659)

  LiuChiachi

  authored
  9adaf58

  Copy full SHA for 9adaf58

Commits on Jan 27, 2022

• [Sharding] Fix gpt3 sharding hybrid parallel (#1648)

  Baibaifan

  authored
  c728cab

  Copy full SHA for c728cab

Commits on Jan 26, 2022

• [GPT-3] Support sharding stage2, stage3 for gpt-3 dygraph training (#1636)

  Show description for c720cf6
  

  Baibaifan

  and

  ZHUI

  authored
  c720cf6

  Copy full SHA for c720cf6

• Fix batchify function of token classification task of FasterErnie. (#1583)

  Show description for 8e94a97
  joey12300

  authored
  8e94a97

  Copy full SHA for 8e94a97

• [BUGFIX] Cnn_dailymail and xnli raise error when downloading in multi-gpus mode (#1587)

  Show description for 2197402
  

  gongenlei

  and

  LiuChiachi

  authored
  2197402

  Copy full SHA for 2197402

Commits on Jan 25, 2022

• [BugFix] [Pre-training] Fix mlm label of dataset_utils (#1633)

  ZHUI

  authored
  82b1cc4

  Copy full SHA for 82b1cc4

• Fix Transformer amp (Fix #1574) (#1635)

  FrostML

  authored
  2455ec2

  Copy full SHA for 2455ec2

Commits on Jan 22, 2022

• Update PPMiniLM performance result tested by QA (#1602)

  Show description for 22a497c
  LiuChiachi

  authored
  22a497c

  Copy full SHA for 22a497c

Commits on Jan 17, 2022

• Update shape range setting method (#1599)

  Show description for 27d87b6
  LiuChiachi

  authored
  27d87b6

  Copy full SHA for 27d87b6

• Fix ppminilm export bug (#1596)

  Show description for 22cee26
  LiuChiachi

  authored
  22cee26

  Copy full SHA for 22cee26

Commits on Jan 15, 2022

• [Pre-Training] Add tutorial for clue small 14g dataset (#1555)

  Show description for a5f8a3e
  

  ZHUI

  and

  ZeyuChen

  authored
  a5f8a3e

  Copy full SHA for a5f8a3e

Commits on Jan 13, 2022

• fix: fix windows dtype to int64 (#1588)

  

  gongenlei

  authored
  60608d3

  Copy full SHA for 60608d3

• Update neural search readme and Add Paddle Serving Support (#1558)

  Show description for 6213573
  

  3 peopleauthored
  6213573

  Copy full SHA for 6213573

• update paddle version in readme when using amp training (#1586)

  FrostML

  authored
  2a620ba

  Copy full SHA for 2a620ba

Commits on Jan 12, 2022

• update ppminilm readme (#1584)

  Show description for e321e09
  

  LiuChiachi

  and

  ZeyuChen

  authored
  e321e09

  Copy full SHA for e321e09

• [CI] Add max_steps for pointer_summarizer and ernie-gen (#1556)

  Show description for b313085
  

  gongenlei

  authored
  b313085

  Copy full SHA for b313085

• Fix PPMiniLM quantization bug (#1580)

  Show description for 85f4cc0
  LiuChiachi

  authored
  85f4cc0

  Copy full SHA for 85f4cc0

• [amp] refine transformer fp16 train (#1574)

  zhangbo9674

  authored
  1e4975d

  Copy full SHA for 1e4975d

Commits on Jan 11, 2022

• Add FasterTokenizer on PPMiniLM (#1542)

  Show description for 4ac2811
  LiuChiachi

  authored
  4ac2811

  Copy full SHA for 4ac2811

• Unified customization for Taskflow (#1517)

  Show description for c1d5241
  linjieccc

  authored
  c1d5241

  Copy full SHA for c1d5241

Commits on Jan 8, 2022

• process data when required avoid occupied too much memory (#1550)

  Show description for e80b4c1
  

  Steffy-zxf

  and

  ZeyuChen

  authored
  e80b4c1

  Copy full SHA for e80b4c1

Commits on Jan 7, 2022

• fix: add args.min_target_length (#1560)

  

  gongenlei

  authored
  be5299e

  Copy full SHA for be5299e

Commits on Jan 4, 2022

• use paddle.einsum instead of paddlenlp.ops.einsum (#1557)

  joey12300

  authored
  3d3bc82

  Copy full SHA for 3d3bc82

Commits on Dec 31, 2021

• Fix FasterTransformer when export float16 model (#1539)

  FrostML

  authored
  e8c7474

  Copy full SHA for e8c7474

Commits on Dec 30, 2021

• Merge pull request #1515 from joey12300/fix_some_error

  Show description for 128a7e5
  wawltor

  authored
  128a7e5

  Copy full SHA for 128a7e5

• fix the ips of Bert in the static graph (#1518)

  Show description for e1842d7
  

  Yanxing-Shi

  and

  ZeyuChen

  authored
  e1842d7

  Copy full SHA for e1842d7

• refine doc

  joey12300

  committed
  05e80c0

  Copy full SHA for 05e80c0

• [ERNIE-M] Add ernie-m (#1510)

  Show description for 7c8dd35
  

  gongenlei

  authored
  7c8dd35

  Copy full SHA for 7c8dd35

Commits on Dec 29, 2021

• Add transformer tipc (#1472)

  Show description for cc9dc32
  

  FrostML

  and

  ZeyuChen

  authored
  cc9dc32

  Copy full SHA for cc9dc32

• [XNLI] Add xnli dataset (#1474)

  Show description for 49f7f0d
  

  gongenlei

  and

  ZeyuChen

  authored
  49f7f0d

  Copy full SHA for 49f7f0d

• Add ppminilm inference (#1503)

  Show description for 032677e
  LiuChiachi

  authored
  032677e

  Copy full SHA for 032677e

Commits on Dec 28, 2021

• Fix return_mask bug in squad for ernie, bert model (#1522)

  Show description for ec7fb3d
  joey12300

  authored
  ec7fb3d

  Copy full SHA for ec7fb3d

• pdate general distill in ppminilm (#1520)

  LiuChiachi

  authored
  d82766b

  Copy full SHA for d82766b

• Add PPMiniLM class (#1512)

  Show description for a16c4bc
  LiuChiachi

  authored
  a16c4bc

  Copy full SHA for a16c4bc