Description

nnet: Theano-based Deep Learning library ## Overview Nnet is a lightweight library to build and train neural networks in [Theano](http://deeplearning.net/software/theano/) and written in Python. The objective of this library is focus on enabling fast experimentation and prototype in deep learning. Features: - Fast prototyping (through total modularity, minimalism, and extensibility). - Runs seamlessly on CPU and GPU. - Freely definable cost function and no need to derive gradients due to Theano's symbolic differentiation - Simplicity ## Guiding principles - __Minimalism.__ Each module should be kept short and simple (<100 lines of code). Every piece of code should be transparent upon first reading. - __Work with Python__. No separate models configuration files in a declarative format (like in Caffe or PyLearn2). Models are described in Python code, which is compact, easier to debug, and allows for ease of extensibility. ## Code Find the code on Github: [kishorenayar/Nnet](https://github.com/kishorenayar/Nnet).