huggingface/transformers

Pull Requests Commits

`torch.distributed` group initialization for `torch_neuron` disabled when `optimum-neuron` is installed (#22728)

michaelbenayoun committed 3 years ago

Verified 10fab90f

[tests] switch to torchrun (#22712)

stas00 committed 3 years ago

Verified 1306b7d3

Modify pipeline_tutorial.mdx (#22726)

ARKA1112 committed 3 years ago

Verified d87ef00c

[`bnb`] Let's make serialization of int8 models possible (#22177)

younesbelkada committed 3 years ago

Verified 370f0ca1

add model resources for CPMAnt (new) (#20906)

pioliverse committed 3 years ago

Verified 523ca4e0

Added parallel device usage for GPT-J (#22713)

jprivera44 committed 3 years ago

Verified 17503b00

remove wrong doc in readme (#22723)

ArthurZucker committed 3 years ago

Verified b76e6ebd

Update input values for docstring (#22631)

amyeroberts committed 3 years ago

Verified 5a71977b

Fix decorator order (#22708)

ydshieh committed 3 years ago

Verified fe1f5a63

Replace -100s in predictions by the pad token (#22693)

sgugger committed 3 years ago

Verified 1b1867d8

Remove 2 failing ONNX conversion tests (#22660)

ydshieh committed 3 years ago

Verified ff73deeb

Clarify stride option (#22684)

luccailliau committed 3 years ago

Verified 06b05d45

Enable naive Pipeline Parallelism training for Gpt neox japanese and san japanese (#22702)

mayankagarwals committed 3 years ago

Verified 0224aaf6

Make it easier to develop without a dev install (#22697)

sgugger committed 3 years ago

Verified 28c19ab5

Update some `MarkupLM` tests' expected values (#22667)

ydshieh committed 3 years ago

Verified 4c01231e

Model parallelism: Moving labels to same devices as the logits are (#22691)

shahad-mahmud committed 3 years ago

Verified 151425dd

add GPTNeoXForSequenceClassification (#22671)

Asugawara committed 3 years ago

Verified 6daa9cb5

use __func__ to check can_generate (#22643)

xin3he committed 3 years ago

Verified f74b4020

Fix quantization docs typo (#22666)

python273 committed 3 years ago

Verified 14fc1a24

Make dynamic code work with offline mode (#22661)

sgugger committed 3 years ago

Verified 3876fc68

(feat): Moving labels to same device as logits for Deit (#22679)

xssChauhan committed 3 years ago

Verified 98597725

Model parallelism: Moving labels to the same device as logits for BridgeTower models (#22676)

shahad-mahmud committed 3 years ago

Verified 870d91fb

Add GPTBigCode model (Optimized GPT2 with MQA from Santacoder & BigCode) (#22575)

jlamypoirier committed 3 years ago

Verified e0921c6b

moved labels to the same device as logits for BLOOM, GPT Neo, GPT NeoX, RoBERTa and VIT models (#22663)

iamarunbrahma committed 3 years ago

Verified 656e869a

Revert migration of setup to pyproject.toml (#22658)

sgugger committed 3 years ago

Verified 6db23af5

Generate: add API warning to streamers (#22659)

gante committed 3 years ago

Verified 3f96e0b4

[OPT] Fix default attention mask size (#22649)

ArthurZucker committed 3 years ago

Verified f3341926

[tokenization] do not push special file (#22657)

ArthurZucker committed 3 years ago

Verified b1b3dc3e

Small nit, (#22653)

ArthurZucker committed 3 years ago

Verified 117a0f6a

🌐 [i18n-KO] Translated `pipeline_tutorial.mdx` to Korean (#22508)

wonhyeongseo committed 3 years ago

Verified fc1ba6fd

Newer Older