Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Mar 24, 2023

• [safetensors] don't use in torch<1.10 (huggingface#22370)

  Show description for cae78c4
  stas00

  authored
  cae78c4

  Copy full SHA for cae78c4

• Fix TF pipeline job

  sgugger

  committed
  cfab34e

  Copy full SHA for cfab34e

• [Trainer] add disclaimer that full_determinism is slow (huggingface#22368)

  stas00

  authored
  500fce0

  Copy full SHA for 500fce0

• Resnet flax (huggingface#21472)

  Show description for a0cbbba
  

  Shubhamai

  and

  amyeroberts

  authored
  a0cbbba

  Copy full SHA for a0cbbba

• TensorFlow: pin maximum version to 2.12 (huggingface#22364)

  gante

  authored
  88dae78

  Copy full SHA for 88dae78

• Improve error message (huggingface#22361)

  Show description for 3a7f5fa
  Mahrkeenerh

  authored
  3a7f5fa

  Copy full SHA for 3a7f5fa

• Pin tensorflow-text to go with tensorflow (huggingface#22362)

  Show description for 6587125
  sgugger

  authored
  6587125

  Copy full SHA for 6587125

• Update docker files to use official torch 2.0.0 (huggingface#22357)

  Show description for 0120347
  ydshieh

  authored
  0120347

  Copy full SHA for 0120347

• Add Mega: Moving Average Equipped Gated Attention (huggingface#21766)

  Show description for 57f25f4
  

  3 peopleauthored
  57f25f4

  Copy full SHA for 57f25f4

• Generate: Add GPTNeoX integration test (huggingface#22346)

  gante

  authored
  0fa4652

  Copy full SHA for 0fa4652

• Fix typo in Greedy Search Description (huggingface#22345)

  Show description for b796076
  awinml

  authored
  b796076

  Copy full SHA for b796076

• [HFTracer] Make embeddings ops take on the dtype of the weight (huggingface#22347)

  Show description for c0fa2aa
  jamesr66a

  authored
  c0fa2aa

  Copy full SHA for c0fa2aa

Commits on Mar 23, 2023

• Automatically create/update tiny models (huggingface#22275)

  Show description for e8cc025
  ydshieh

  authored
  e8cc025

  Copy full SHA for e8cc025

• Enable training Llama with model or pipeline parallelism (huggingface#22329)

  Show description for a92e0ad
  

  kooshi

  and

  sgugger

  authored
  a92e0ad

  Copy full SHA for a92e0ad

• Generate: add test for left-padding support (huggingface#22322)

  gante

  authored
  502fec7

  Copy full SHA for 502fec7

• Fix --bf16 option support for Neuron after PR huggingface#22300 (huggingface#22307)

  Show description for ec9b18f
  jeffhataws

  authored
  ec9b18f

  Copy full SHA for ec9b18f

• Added type hints to TFDeiTModel (huggingface#22327)

  Show description for aef488c
  

  Batese2001

  and

  Rocketknight1

  authored
  aef488c

  Copy full SHA for aef488c

• Minor typo in pipeline FillMaskPipeline's documentation. (huggingface#22339)

  SamuelLarkin

  authored
  59b9351

  Copy full SHA for 59b9351

• Fix various imports (huggingface#22281)

  Show description for 506e7c6
  sgugger

  authored
  506e7c6

  Copy full SHA for 506e7c6

• Mention why one needs to specify max_steps in Trainer (huggingface#22333)

  Show description for 053c215
  lhoestq

  authored
  053c215

  Copy full SHA for 053c215

• Fixed gradient checkpoint bug for TimeSeriesTransformer (huggingface#22272)

  Show description for 5a9eb31
  

  pmollerus23

  and

  younesbelkada

  authored
  5a9eb31

  Copy full SHA for 5a9eb31

• [MBart] Add accelerate support for MBart (huggingface#22309)

  Show description for ff20f9c
  younesbelkada

  authored
  ff20f9c

  Copy full SHA for ff20f9c

• [gptj] support older pytorch version (huggingface#22325)

  Show description for 61f79b2
  

  3 peopleauthored
  61f79b2

  Copy full SHA for 61f79b2

• Really fix quality due to ruff release

  sgugger

  committed
  80e3b36

  Copy full SHA for 80e3b36

• Fix quality due to ruff release

  sgugger

  committed
  ef28df0

  Copy full SHA for ef28df0

Commits on Mar 22, 2023

• [deepspeed zero3] need generate(synced_gpus=True, ...) (huggingface#22242)

  Show description for 73fdc8c
  

  stas00

  and

  sgugger

  authored
  73fdc8c

  Copy full SHA for 73fdc8c

• Fix PipelineTests skip conditions (huggingface#22320)

  Show description for 8b05ace
  ydshieh

  authored
  8b05ace

  Copy full SHA for 8b05ace

• Chunkable token classification pipeline (huggingface#21771)

  Show description for d62e7d8
  

  3 peopleauthored
  d62e7d8

  Copy full SHA for d62e7d8

• docs: Resolve incorrect type typo in trainer methods (huggingface#22316)

  Show description for f48d331
  tomaarsen

  authored
  f48d331

  Copy full SHA for f48d331

• Add Pix2Struct (huggingface#21400)

  Show description for 0f68a7f
  

  4 peopleauthored
  0f68a7f

  Copy full SHA for 0f68a7f

• Beef up Llama tests (huggingface#22314)

  Show description for fd3eb3e
  gante

  authored
  fd3eb3e

  Copy full SHA for fd3eb3e

• Generate: Export TF generate with a TF tokenizer (huggingface#22310)

  Show description for 12febc2
  gante

  authored
  12febc2

  Copy full SHA for 12febc2

• Enforce max_memory for device_map strategies (huggingface#22311)

  Show description for 5fd4e3c
  sgugger

  authored
  5fd4e3c

  Copy full SHA for 5fd4e3c

• Fixed bug to calculate correct xpath_sub_list in MarkupLMTokenizer (huggingface#22302)

  Show description for 48bef3a
  

  silentghoul-spec

  and

  dusejat

  authored
  48bef3a

  Copy full SHA for 48bef3a

• Fix position embeddings for GPT-J and CodeGen (huggingface#22069)

  Show description for 4e94c6c
  njhill

  authored
  4e94c6c

  Copy full SHA for 4e94c6c

Pagination

Previous

Next