https://github.com/athemathmo/vision-rs

Access to computer vision benchmarking datasets in Rust
https://github.com/athemathmo/vision-rs

Last synced: 3 days ago
JSON representation

Access to computer vision benchmarking datasets in Rust

• Host: GitHub
• URL: https://github.com/athemathmo/vision-rs
• Owner: AtheMathmo
• Created: 2018-02-27T17:25:11.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-02-27T17:51:35.000Z (over 6 years ago)
• Last Synced: 2024-08-10T07:37:22.109Z (3 months ago)
• Language: Rust
• Size: 5.86 KB
• Stars: 6
• Watchers: 3
• Forks: 1
• Open Issues: 3
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Vision

This library provides access to common machine learning benchmarking datasets.

The library currently includes:

- [MNIST](http://yann.lecun.com/exdb/mnist/)

- [FashionMNIST](https://github.com/zalandoresearch/fashion-mnist)

- [CIFAR-10](https://www.cs.toronto.edu/~kriz/cifar.html)

- [CIFAR-100](https://www.cs.toronto.edu/~kriz/cifar.html)

Things are currently _very_ basic.

## Usage

Each dataset can be downloaded and processed using a Builder class. The builder is customizable in each case.

```rust

extern crate vision;

use vision::mnist::{MNISTBuilder};

fn main() {

    let builder = MNISTBuilder::new();

    let mnist = builder.data_home("MNIST")

                       .verbose()

                       .get_data().unwrap();

    println!("{}", mnist.train_imgs.len());

}

```

The MNIST object returned by the builder contains four public fields, `train_imgs`, `train_labels`, `test_images` and `test_labels`. The label fields are `Vec` types and the images are `Vec>`, each entry in the outermost `Vec` corresponds to a single datapoint.

Further preprocessing should be carried out by the user.