edaviz
Python Library for Data Exploration and Visualization in Jupyter Notebook and Jupyter Lab
Watch the demo video.
Currently, edaviz is in private alpha. If you like edaviz, you can subscribe to our waiting list in order to get early access.
If you already have early access, you can find the documentation here.
Outline
The Problem
80% of the effort in Data Science projects is due to data understanding, transformation, and cleaning. And of course, it all starts with data exploration. If you don't understand the data set, you cannot work with it.
However, the path to understanding your data is paved with numerous tedious tasks. You need to investigate the distribution of your columns, explore missing values, and check whether there are some useful patterns in your data. In most cases, we need to be cautious with our data, as there are always odd suprises to be found.
The Solution
edaviz is a Python library which helps you to quickly understand and visualize your data frames. It is most helpful during data exploration after you received a new data set. However, you can also use it during data cleaning and data transformation, during feature engineering, or during model evaluation - whenever you need to understand your data.
Watch the demo video
The Features
- Quick exploration without coding but full flexibility of code (after all, you are working in a Jupyter environment)
- High interactivity and data drill down
- Finds the best features for predicting your target
- Explores predictive patterns between columns
- Automatically chooses visualizations based on data types
Watch the demo video
Contact
If you have questions or want to contribute, you can contact us via opening a GitHub issue or texting us on twitter.
