/PyPatent

Measures similarity between patents and publications that cite them via doc2vec

Primary LanguagePythonMIT LicenseMIT

PyPatent

File structure

  • PyPatent (dir)
    • main.py
    • readpatentpdf.py
    • readabstracttxt.py
    • TEXT_Files (dir containing initial data)
      • Patent 1 & literature searches (dir)
      • ...
      • Patent N & literature searches (dir)
  • Train (dir containing pre-processed data)
    • abstract2vec.py
    • a2v.d2v (the trained model)
    • training txt files converted from patents and abstracts

How to run PyPatent?

  1. Clone or fork this repository
  2. Create a Python enviornment with the repo's requirements.txt libraries (e.g. conda create -n pypatent python=3.5 anaconda --file requirements.txt)
  3. Activate that enviornment (source activate pypatent)
  4. Run main.py from the project directory((pypatent) computer:PyPatent hclent$ python main.py)
    • This will produce a results.csv file, with the results.

Debugging?

After you run the code once, there will be a .pypatent.log file, which an help you if anything has gone amiss.