SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Deprecate low use models (#30781)

Browse Source 
* Deprecate models
- graphormer
- time_series_transformer
- xlm_prophetnet
- qdqbert
- nat
- ernie_m
- tvlt
- nezha
- mega
- jukebox
- vit_hybrid
- x_clip
- deta
- speech_to_text_2
- efficientformer
- realm
- gptsan_japanese

* Fix up

* Fix speech2text2 imports

* Make sure message isn't indented

* Fix docstrings

* Correctly map for deprecated models from model_type

* Uncomment out

* Add back time series transformer and x-clip

* Import fix and fix-up

* Fix up with updated ruff
This commit is contained in:
amyeroberts
2024-05-28 18:07:07 +01:00
committed by GitHub
parent 7f08817be4
commit a564d10afe
142 changed files with 1308 additions and 11908 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
docs/source/en/model_doc/mega.md
View File

@@ -16,12 +16,20 @@ rendered properly in your Markdown viewer.
# MEGA
<Tip warning={true}>
This model is in maintenance mode only, we don't accept any new PRs changing its code.
If you run into any issues running this model, please reinstall the last version that supported this model: v4.40.2.
You can do so by running the following command: `pip install -U transformers==4.40.2`.
</Tip>
## Overview
The MEGA model was proposed in [Mega: Moving Average Equipped Gated Attention](https://arxiv.org/abs/2209.10655) by Xuezhe Ma, Chunting Zhou, Xiang Kong, Junxian He, Liangke Gui, Graham Neubig, Jonathan May, and Luke Zettlemoyer.
MEGA proposes a new approach to self-attention with each encoder layer having a multi-headed exponential moving average in addition to a single head of standard dot-product attention, giving the attention mechanism
stronger positional biases. This allows MEGA to perform competitively to Transformers on standard benchmarks including LRA
while also having significantly fewer parameters. MEGA's compute efficiency allows it to scale to very long sequences, making it an
MEGA proposes a new approach to self-attention with each encoder layer having a multi-headed exponential moving average in addition to a single head of standard dot-product attention, giving the attention mechanism
stronger positional biases. This allows MEGA to perform competitively to Transformers on standard benchmarks including LRA
while also having significantly fewer parameters. MEGA's compute efficiency allows it to scale to very long sequences, making it an
attractive option for long-document NLP tasks.
The abstract from the paper is the following:
@@ -34,8 +42,8 @@ The original code can be found [here](https://github.com/facebookresearch/mega).
## Usage tips
- MEGA can perform quite well with relatively few parameters. See Appendix D in the MEGA paper for examples of architectural specs which perform well in various settings. If using MEGA as a decoder, be sure to set `bidirectional=False` to avoid errors with default bidirectional.
- Mega-chunk is a variant of mega that reduces time and spaces complexity from quadratic to linear. Utilize chunking with MegaConfig.use_chunking and control chunk size with MegaConfig.chunk_size
- MEGA can perform quite well with relatively few parameters. See Appendix D in the MEGA paper for examples of architectural specs which perform well in various settings. If using MEGA as a decoder, be sure to set `bidirectional=False` to avoid errors with default bidirectional.
- Mega-chunk is a variant of mega that reduces time and spaces complexity from quadratic to linear. Utilize chunking with MegaConfig.use_chunking and control chunk size with MegaConfig.chunk_size
## Implementation Notes