Skip to main content

SQUAT

Project description

Spend Quality and Usage Analysis Tool (SQUAT)

This Project is a tool to analyse Bankstatements transactions to give a comprehensive report on the spend, earning and usage of an user. It does the following job:

  • Creates and trains a Machine learning model to classify transactions based on the narration.
  • All the training and other repeatative work is already done for you.
  • Once the package is installed with pip, the developer just need to pass the bankstatement dataframe to get the report.



Project Components

SQUAT contains the packages or libraries required for supporting and running the whole process.

  1. spacy
  2. Core ENG package for spacy
  3. pandas
  4. jupyter notebook

Source:
https://github.com/binayr/SQUAT

About the ML model

The model is created based on most common keyword observed from the bankstatements of singapore. This project has a large scope of improving the accuracy and adding more classifications in future depending on the type of dataset available to us.

Everytime we update the model a new version of SQUAT is supposed to get released.

Create and use whl file

  • with and updated setup.py execute the following command to create a whl file, python setup.py bdist_wheel

  • Please make sure you have pre-installed pandas, spacy and jupyter from standard chartered artifactory in your virtualenv

  • Also make sure once spacy is installed the eng core library is also pre-installed in the virtualenv using pip.

  • Now you can pip install squat using the whl file or from standard chartered artifactory if it is hosted.

API

  • You can import the utility by typing the following, from squat.Classifier.ClassifierUtil import ClassifierUtil

  • Read any csv or excel using pandas and create a dataframe. Please make sure the df has the following header atleast, date, description, debit, credit, runningbalance (irrespective of the order)

  • The ClassifierUtil can be initialized using the above df.

  • Once initialized please make sure to call obj.evaluate() to evaluate each transaction.

  • Once evaluated you can call get_analysis method to get the comprehensive analysis or call show_stat to get the statistics.

OR

  • You can import the utility by typing the following, from squat.Classifier.ClassifierUtil import ClassifierUtilRaw

  • Read any csv or excel using pandas and create a dataframe. Please make sure the df has the following header atleast, date, description, debit, credit, runningbalance (irrespective of the order)

  • The ClassifierUtilRaw can be initialized to get the category.

  • Once initialized please make sure to call obj.get_cat(text) to evaluate the category of the text.

  • For Example,

     obj.get_cat('paytm transaction gurgaon')
     Out: ('Digital', 0.9632782936096191)
    

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for squat, version 1.0.3
Filename, size File type Python version Upload date Hashes
Filename, size squat-1.0.3-py3-none-any.whl (3.5 MB) File type Wheel Python version py3 Upload date Hashes View hashes
Filename, size squat-1.0.3.tar.gz (3.5 MB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page