easynn 0.1.3-alpha

(Pre-alpha) A rust crate for quick-and-dirty neuro network developing and training.
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
# easynn

This crate aims to provide neuro network developing and training utilities
in rust, where a variety of models and layers are supported.

Note: this crate is currently in pre-alpha, any interface is subject to change.

### Supported models
 - [ ] `Sequential`: similar to [The Sequential model]https://www.tensorflow.org/guide/keras/sequential_model of [Keras]https://keras.io/

### Supported layer types
 - Primitive types:
   - [x] `Dense`: fully connected layers
 - CNN types:
   - [ ] `Conv`: the convolution layer
   - [ ] `Pooling`: the pooling layer

License: MIT