Skip to content

Testing codes for "One-Shot Image Classification by Learning to Restore Prototypes"

License

Notifications You must be signed in to change notification settings

xuewanqi/RestoreNet

Repository files navigation

RestoreNet

Testing codes for "One-Shot Image Classification by Learning to Restore Prototypes"

The repository contains the essential codes for RestoreNet and hope it will help to understand the idea.

  1. You can do your own training, or alternatively, you may like to download the network parameters and pre-computed features at https://www.dropbox.com/sh/6jy0g8nfc97bvrm/AACORpPowNVnFXdwek7vUYjIa?dl=0

    In the link, there are three files: 1) FeatureExtractor.pth 2) Transformer.pth 3) GalleryPool.

    The first two .pth files are network parameters. Please put them in the folder ./Network Params/

    The GalleryPool file is pre-computed images features used for self-training. We use it to avoid unnecessary computation. Please put it in ./

  2. Put all images in miniImageNet dataset in the folder /miniImageNet/images.

  3. Do evaluation by run 'python run_test.py'

About

Testing codes for "One-Shot Image Classification by Learning to Restore Prototypes"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages