Downloads · 30 days
277
1% of all-time downloads
sonoisa/clip-vit-b-32-japanese-v1
clip-vit-b-32-japanese-v1 is a feature extraction model from sonoisa. Use it when you need embeddings to search or compare text. It is set up for transformers. The card lists the license as cc-by-sa-4.0.
This is a CLIP text/image encoder model for Japanese.
Downloads · 30 days
277
1% of all-time downloads
All-time downloads
26K
Public
Parameters
111M
2.6 GB on disk
Likes
21
Public
Click a slice to open those files.
.bin803 MB · 64%
How the weights are stored.
F32111M · 100%
From the Hugging Face model README
This is a CLIP text/image encoder model for Japanese.
英語版CLIPモデルのテキストエンコーダーを一種の蒸留を用いて日本語化したモデルです。 作り方や精度、使い方、サンプルコードは下記の解説記事をご参照ください。
解説記事:
サンプルコードのリポジトリ: https://github.com/sonoisa/clip-japanese
デモ: