Skip to content
Applying XBRL to AI
HTML Python Jupyter Notebook
Branch: master
Clone or download
Niels-Peter Rønmos
Niels-Peter Rønmos update_new_dk_url
Latest commit e18a9b1 Nov 7, 2018
Type Name Latest commit message Commit time
Failed to load latest commit information.
documents simple example Mar 12, 2018
example issue May 8, 2018
xbrl_local update_new_dk_url Nov 7, 2018
.gitignore Initial commit Jan 30, 2018
LICENSE Initial commit Jan 30, 2018 new file structure Feb 9, 2018
_config.yml Set theme jekyll-theme-modernist Feb 9, 2018
environment.yml Removed unnecessary channels Feb 3, 2018 samples Feb 13, 2018 Belgium first version Apr 25, 2018


The purpuse off this project is to apply XBRL in AI and Machine learning, by usage of Python.

The project will be divided into 3 parts:

  1. Generic XBRL to AI.
  1. XBRL in local GAAP including
  • Danish GAAP og Danish IFRS (with extension) to AI
  • German GAAP
  1. Sample of Machine Learning implementation based on this project

Why this project?

Working with machine learning basicly comes down to one thing: y = f(X), y is what we want to predict, X is the input and f is the machine learning model. Unfortunately X hardly ever fits into to f. If we want to fit e.g. an XBRL-instance into to f, we need to prepare the data. XBRL needs a good representation to fit into AI and Machine learning.

Creating a good and standardized representation of XBRL into AI and Machine learning are the main purpose of this project.

Getting started

To see an example of how one could use xbrl-ai start by creating a conda environment from the cloned yaml file:

>> conda env create -f environment.yml

now activate the environment by using

>> activate xbrl_ai

for windows, or

>> source activate xbrl_ai

for linux.

You can now run the


Download the files and run the following command from the folder:

  • Using pip:
>> pip install .
  • Using setuptools:
>> python install
You can’t perform that action at this time.