Image captioning on Menoh
The model is based on Chainer's example.
$ git clone https://github.com/Hakuyume/menoh-captioning.git
$ cd menoh-captioning
$ curl -LO https://github.com/Hakuyume/menoh-captioning/releases/download/assets/model.npz
$ curl -LO https://github.com/Hakuyume/menoh-captioning/releases/download/assets/vocab.txt
$ python3 convert.py model.npz
$ curl -LO https://upload.wikimedia.org/wikipedia/commons/7/79/Trillium_Poncho_cat_dog.jpg
$ cargo run --release -- Trillium_Poncho_cat_dog.jpg
a dog laying on a bed with a blanket .
$ docker run -it hakuyume/menoh-captioning menoh-captioning Trillium_Poncho_cat_dog.jpg
a dog laying on a bed with a blanket .
Content type
Image
Digest
Size
932.3 MB
Last updated
almost 8 years ago
docker pull hakuyume/menoh-captioning