The open model hub is a growing collection of data models that you can take, combine and run for product analysis and exploration.
Project description
Open model hub
The open model hub is a toolkit with functions and models that can run directly on a full dataset collected with Objectiv's Tracker SDKs. All models are open-source, free to use, and can easily be combined to build advanced compound models.
How to use the open model hub
The following types of functions/models are provided:
- Helper functions: simplify manipulating and analyzing the data.
- Aggregation models: enable running some of the more common data analyses and product analytics metrics.
- Machine learning models.
- Funnels: To analyze Funnels, e.g. discover all the (top) user journeys that lead to conversion or drop-off.
Modeling behavior of users and groups is enabled through configurable Identity Resolution.
See how to get started in your notebook and the example notebooks, and install the model hub package directly from PyPI:
pip install objectiv-modelhub
Powered by Bach
The open model hub is powered by Bach: Objectiv’s data modeling library. With Bach, you can compose models with familiar Pandas-like dataframe operations that use an SQL abstraction layer to run on the full dataset. Models can be output to SQL with one command.
Example notebooks
We have some example notebooks in the /notebooks/ directory of the repository that demonstrate how you can work with the data in Python. These notebooks can run on your collected data. The only thing that might need to be adjusted is how the connection to the database is made (see below). All other instructions live in the README.md in the link above.
Set up development environment
This section is only required for development on the objectiv-modelhub package.
virtualenv venv
source venv/bin/activate
export PYTHONPATH=.
# Install both Bach and ModelHub in local 'edit' mode
pip install -e ../bach
pip install -e .[dev]
# The above will fail if the postgres lib development headers are not present. On Ubuntu that can be fixed
# with: sudo apt-get install libpq-dev
PyCharm
- Mark the following directories as "Sources root":
/bach/
/modelhub/
- Set
modelhub/venv/bin/python
as the default interpreter for the project
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for objectiv_modelhub-0.0.21-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | af278224d22f50155898d66e05607d96729742defc6ce9f0910bab40b5b67cd7 |
|
MD5 | aa88587706596764dd14271fa664b298 |
|
BLAKE2b-256 | cb50c01befaec59b1ecfd535351c1973f47f547675b9bcb6422829d8590f22fb |