NLP_PL

Contents

CorpusPopulate

Contains script to retrieve Python Files from /git_dl

Data

Files containing extracted data

Java

Our corpus

LM

The main code:

ExtractCode.py - CodeUnit / Code analysis ExtractSequences.py - Generate sequences from CodeUnits LM.py - The Language Model ProcessData.py - Gets info for LM TestModel.py - The actual test TypeUtils.py - misc. WriteData.py - Processes CodeUnits results.csv - THE LL's

git_dl

Store zip files here

Process

Build Corpus

DL .zip files to git_dl CorpusPopulate/GetJavaFiles.py

Build LM

mode = {levels, cfs} WriteData.py mode ProcessData.py mode

Run LM

TestModel.py