A basic deep learning framework written in Rust, allowing easy definition of Back-Propagation Network. Based on the static computational graph, this framework has a decent running speed on the CPU. You can see example of mnist dataset at examples/mnist.rs. To run the example, you need to download the mnist dataset from here (if it doesn't work, try here). After downloading all the 4 files, extract and put them in a new folder named data under the root of the project (if you are downloading from kaggle, extract the four files from the four folders). Then run the example with cargo run --example mnist --release.
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|