NLP Architect by Intel® AI Lab¶
NLP Architect is an open-source Python library for exploring the state-of-the-art deep learning topologies and techniques for natural language processing and natural language understanding. It is intended to be a platform for future research and collaboration.
The library includes our past and ongoing NLP research and development efforts as part of Intel AI Lab.
NLP Architect can be downloaded from Github: https://github.com/NervanaSystems/nlp-architect
Research driven NLP/NLU models¶
The library contains state-of-art and novel NLP and NLU models in a variety of topics:
Intent detection and Slot tagging model for Intent based applications
Memory Networks for goal-oriented dialog
Noun phrase embedding vectors model
Noun phrase semantic segmentation
Named Entity Recognition
Language modeling using Temporal Convolution Network
Unsupervised Crosslingual Word Embedding
Aspect Based Sentiment Analysis
Supervised sentiment analysis
Sparse and quantized neural machine translation
Relation Identification and cross document coreference
Select the desired configuration of your system:
|Install in developer mode?|
Run the following commands to install NLP Architect:
It is recommended to install NLP Architect in development mode to utilize all its features, examples and solutions.
How can NLP Architect be used¶
Train models using provided algorithms, reference datasets and configurations
Train models using your own data
Create new/extend models based on existing models or topologies
Explore how deep learning models tackle various NLP tasks
Experiment and optimize state-of-the-art deep learning algorithms
integrate modules and utilities from the library to solutions
Deep Learning frameworks¶
Because of the current research nature of the library, several open source deep learning frameworks are used in this repository including:
Overtime the list of models and frameworks included in this space will change, though all generally run with Python 3.6+
Using the Models¶
Each of the models includes a comprehensive description on algorithms, network topologies, reference dataset descriptions and loader, and evaluation results. Overtime the list of models included in this space will grow.