cifar-ten
Parses the binary files of the CIFAR-10 data set and returns them as a pair of tuples (data, labels) with of type and dimension:
- Training data:
Array4<u8> [50_000, 3, 32, 32]andArray2<u8> [50_000, 10] - Testing data:
Array4<u8> [10_000, 3, 32, 32]andArray2<u8> [10_000, 10]
OR
- as a set of flattened
Array2<f32>structures in the same arrangement.
A random image from each dataset and the associated label can be displayed upon parsing. A tar.gz file with the original binaries can be found here. This can be download manually, or automatically using the download feature.
use *;
To add as a dependency, please use:
[]
= "0.1"
Note: Previous commits have included the dataset, which will make the download size large. For development, it's suggested to
$ git clone --depth=1 https://github.com/quietlychris/cifar-ten.git