/DRL4IOT

NLP DRL System for learning controls from JSON, XML and other data formats without manual pre-processing

Primary LanguagePythonOtherNOASSERTION

DRL4IOT scap-2020-code-release

This is the supplemental material to the paper "DRL for IoT Interoperability" found on: https://www.researchgate.net/publication/341641724_Deep_Reinforcement_Learning_for_IoT_Interoperability

Setup

Follow these steps to setup the repository:

  1. Create a new virtual env using python 3.6
  2. Install requirements:
    pip install -r requirements.txt
  3. Install NLTK data:
    python -c "import nltk; nltk.download('punkt')"
  4. Download GloVe pre-trained word vectors and extract in weights directory:
    Linux:
    wget -c http://nlp.stanford.edu/data/wordvecs/glove.6B.zip -P weights
    unzip weights/glove.6B.zip -d weights

Model training

Step by Step training example can be found in Training.ipynb notebook.