For each update, we update this page with a summary of recent significant changes to RATH, including a timestamp for the updated date.
Refer to CHANGELOG.md (opens in a new tab) on our GitHub for a complete record of all historical releases.
- Added BigQuery and Snowflake support as the data source.
- Added README for pygwalker, which allows you to use Graphic Walker within Python Jupyter Notebook.
For the latest update, RATH introduces an excitingly powerful tool for Causal Analysis.
Causal analysis could be defined as the way to identify and examine the causal relationship between variables, which helps create better prediction models and decision-making.
RATH's causal analysis update includes the following key features:
- Causal Discovery: automatically generates causal models from a dataset.
- Edit your graphical causal models in an editor and input predefined background knowledge for RATH.
- Examine and verify your hypothesis.
- Combine EDA(Exploratory Data Analysis) tools to explore your causal model.
- Using interactive visualization to help you understand causal effects.
- Deploy causal-based machine learning models and test strategies for your casual models.
- Manually edit relationship graphs for causal analysis models.
- Conduct What-if types of causal analysis.
For more instructions about conducting causal analysis with RATH, refer to the Causal analysis chapter.