Neural Modules (NeMo) is a framework-agnostic toolkit for building AI applications powered by Neural Modules. Current support is for PyTorch framework.
A "Neural Module" is a block of code that computes a set of outputs from a set of inputs.
Neural Modules’ inputs and outputs have Neural Type for semantic checking.
An application built with NeMo application is a Directed Acyclic Graph(DAG) of connected modules enabling researchers to define and build new speech and nlp networks easily through API Compatible modules.
Documentation and Tutorials
Please refer to the HTML documentation in the docs folder
VIDEO
A Short VIDEO walk-through about using NEMO to experiment with ASR systems.
Core Concepts and Features
- NeuralModule class - represents and implements a neural module.
- NmTensor - represents activation flow between neural modules' ports.
- NeuralType - represents types of modules' ports and NmTensors.
- NeuralFactory - to create neural modules and manage training.
- Lazy execution - when describing activation flow between neural modules, nothing happens until an "action" (such as optimizer.optimize(...) is called.
- Collections - NeMo comes with collections - related group of modules such as nemo_asr (for Speech Recognition) and nemo_nlp for NLP
Requirements
- Python 3.6 or 3.7
- Pytorch >=1.0 with GPU support
- NVIDIA APEX: https://github.com/NVIDIA/apex
- (for nemo_asr do: apt-get install libsndfile1)
Unittests
Run this:
./reinstall.sh
python -m unittest tests/*.pyGetting started
- Clone the repository and run unittests
- Go to nemo folder and do: python setup.py install
- Install collections:
- ASR collection from collections/nemo_asr do: python setup.py install
- NLP collection from collections/nemo_nlp do: python setup.py install
- For development do: python setup.py develop instead of python setup.py install in Step (3) above
- Go to examples/start_here to get started with few simple examples
- To get started with speech recognition:
- head to the ASR tutorial in the documentation
- head to examples/asr/ASR_made_simple.ipynb