WEKO3
Item
Fully Quantized Matrix Arithmetic-Only BERT model (FQ MA-BERT)
https://aist.repo.nii.ac.jp/records/2003355
https://aist.repo.nii.ac.jp/records/20033555eaaeb4a-0a54-48e5-bb87-ed524ccc3249
| Name / File | License | Actions |
|---|---|---|
README.txt (1.6KB)
|
.png)
|
|
src.tar.gz (32KB)
|
|
|
|
rte.tar.gz (387MB)
|
|
|
|
mnli.tar.gz (387MB)
|
|
|
|
cola.tar.gz (387MB)
|
|
| Item type | Research Data v10(1) | |||||
|---|---|---|---|---|---|---|
| PubDate | 2025-05-28 | |||||
| Data name | ||||||
| Title | Fully Quantized Matrix Arithmetic-Only BERT model (FQ MA-BERT) | |||||
| Language | en | |||||
| Description of data | ||||||
| Description Type | Abstract | |||||
| Description | We developed a fully quantized matrix arithmetic-only BERT model (FQ MA-BERT). This repository includes Python codes for the model and trained model parameters. |
|||||
| Language | en | |||||
| Author (Creator) name |
更田, 裕司
× 更田, 裕司 |
|||||
| Terms of Use | ||||||
| Language | en | |||||
| Rights Resource | https://creativecommons.org/licenses/by-nc/4.0/deed.en | |||||
| Rights | Creative Commons Attribution-NonCommercial 4.0 International | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | 自然言語処理 | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | ニューラルネットワーク | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | 深層学習 | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | トランスフォーマー | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | 量子化 | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | BERT | |||||
| Keyword | ||||||
| Language | ja | |||||
| Subject Scheme | Other | |||||
| Subject | Hardware-Friendly | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | Natural Language Processing | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | Neural Network | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | Deep Learning | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | Transformer | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | Quantization | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | BERT | |||||
| Keyword | ||||||
| Language | en | |||||
| Subject Scheme | Other | |||||
| Subject | Hardware-Friendly | |||||
| Publisher | ||||||
| Publisher | National Institute of Advanced Industrial Science and Technology (AIST) | |||||
| Language | en | |||||
| Date | ||||||
| Date | 2025 | |||||
| Date Type | Issued | |||||
| Language | ||||||
| Language | eng | |||||
| Resource Type | ||||||
| Resource Type Identifier | http://purl.org/coar/resource_type/c_ddb1 | |||||
| Resource Type | dataset | |||||
| Identifier | ||||||
| Identifier | https://doi.org/10.57765/2003355 | |||||
| Identifier Type | DOI | |||||
