Downloads · 30 days
20
3% of all-time downloads
albert-embedding-ja
tzmtwtr/albert-embedding-ja
albert-embedding-ja is a sentence similarity model from tzmtwtr. Use it when you need a score for how close two texts are. It is set up for sentence-transformers.
All-time downloads
732
Public
Repo size
95.5 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin47.8 MB · 99%
At a glance
- Task
- Sentence Similarity
- Library
- sentence-transformers
- Model type
- albert
- Access
- Public
- Created
- Jun 13, 2023
- Updated
- Jun 13, 2023
- SHA
- f1e924bf
Model card
From the Hugging Face model README
日本語のSentence Embedding用モデル
以下のモデルから転移学習を実施。
https://huggingface.co/ken11/albert-base-japanese-v1-with-japanese-tokenizer
学習データには以下を使用。
https://huggingface.co/datasets/tzmtwtr/tw-posts-ja
モチベーション
ベクトル検索のために小規模言語モデルが必要になった。
AWS Lambdaで動かせるようにしたい。
- Task
- Sentence Similarity
- Library
- sentence-transformers
- Type
- albert
- Created
- Jun 13, 2023
- Updated
- Jun 13, 2023