Notebooks demonstrating the ModelSkill model validation library
Quantitative skill assessment (validation) of a numerical model is important for most modelling projects. But without proper tools, this can be a time-consuming task. That's why DHI developed ModelSkill, an open-source Python library designed to streamline the skill assessment process for any type of model (i.e., MIKE, FEFLOW, WEST, etc.).
This quick-to-install library offers a wide range of common metrics, simplifies data analysis and provides easy access to plots typically used for model validation (time-series, scatter, Q-Q, etc.). Since ModelSkill is built with Python, it is also customisable and extensible, allowing for the unique formatting of plots and domain-specific metrics if necessary.
This webinar will explore the following topics:
• Getting started with ModelSkill
• Running multi-observation-multi-model comparisons with ease
• Assessing the skill on a subset of data efficiently
• Conducting skill assessment by month
• Showcasing model validation from various domains