subseq 1.0.0

Succinct BWT-Based SequencePrediction

Succinct BWT-Based Sequence Prediction (Subseq)

What is it ?

This project is a c++ implementation with a python wrapper of the Succinct BWT-Based Sequence Prediction model.

Subseq is a sequence prediction model in a finite alphabet. It is a lossless model (does not discard information while training) and utilizes the succinct Wavelet Tree data structure and the Burrows-Wheeler Transform to compactly store and efficiently access training sequences for prediction.

This implementation is based on the following research paper:

• https://www.researchgate.net/publication/335231805_Succinct_BWT-Based_Sequence_Prediction

Installation

Installation from the sources. On Linux and mac, you can easilly install it via make build.

For Windows users, the installation is trickier.

Simple example

You can test the model with the following code:

from subseq.subseq import Subseq
model = Subseq(1)

model.fit([['hello', 'world']])

model.predict(['hello'])
# Output: ['world']

Features

Train

The model can be trained with the fit method.

Tuning

Subseq has only 1 meta parameter that need to be tuned. threshold_query, the number of similar queries that needs to be retrieved to make a confident prediction.

A threshold_query at 0 does not limit the number of query.

Benchmark

The benchmark has been made on the FIFA dataset, the data can be found on the SPMF website.

Details on the benchmark can be found here.