modelskill 1.0a1

Compare results from simulations with observations.

ModelSkill: Flexible Model skill evaluation.

ModelSkill is a python package for scoring MIKE models (other models can be evaluated as well).

Read more about the vision and scope. Contribute with new ideas in the discussion, report an issue or browse the API documentation. Access observational data (e.g. altimetry data) from the sister library WatObs.

Use cases

ModelSkill would like to be your companion during the different phases of a MIKE modelling workflow.

• Model setup - exploratory phase
• Model calibration
• Model validation and reporting - communicate your final results

Installation

From pypi:

> pip install modelskill

Or the development version:

> pip install https://github.com/DHI/modelskill/archive/main.zip

Example notebooks

• Quick_and_dirty_compare.ipynb
• SW_DutchCoast.ipynb
• Multi_model_comparison.ipynb
• Multi_variable_comparison.ipynb
• Track_comparison.ipynb (including global wave model example)
• Spatial_skill.ipynb (satellite tracks, skill aggregated on spatial bins)
• NetCDF_ModelResult.ipynb
• Combine_comparers.ipynb

Workflow

1. Define ModelResults
2. Define Observations
3. Connect Observations and ModelResults
4. Extract ModelResults at Observation positions
5. Do plotting, statistics, reporting using a Comparer

Read more about the workflow in the getting started guide.

Example of use

Start by defining model results and observations:

>>> import modelskill as ms
>>> mr = ms.ModelResult("HKZN_local_2017_DutchCoast.dfsu", name="HKZN_local", item=0)
>>> HKNA = ms.PointObservation("HKNA_Hm0.dfs0", item=0, x=4.2420, y=52.6887, name="HKNA")
>>> EPL = ms.PointObservation("eur_Hm0.dfs0", item=0, x=3.2760, y=51.9990, name="EPL")
>>> c2 = ms.TrackObservation("Alti_c2_Dutch.dfs0", item=3, name="c2")

Then, connect observations and model results, and extract data at observation points:

>>> cc = ms.compare([HKNA, EPL, c2], mr)

With the comparer object, cc, all sorts of skill assessments and plots can be made:

>>> cc.skill().round(2)
               n  bias  rmse  urmse   mae    cc    si    r2
observation                                                
HKNA         385 -0.20  0.35   0.29  0.25  0.97  0.09  0.99
EPL           66 -0.08  0.22   0.20  0.18  0.97  0.07  0.99
c2           113 -0.00  0.35   0.35  0.29  0.97  0.12  0.99

Overview of observation locations

ms.plot_spatial_overview([HKNA, EPL, c2], mr, figsize=(7,7))

Scatter plot

cc.plot.scatter()

Timeseries plot

Timeseries plots can either be static and report-friendly (matplotlib) or interactive with zoom functionality (plotly).

cc["HKNA"].plot.timeseries(width=1000, backend="plotly")

Automated reporting

With a few lines of code, it will be possible to generate an automated report.

from modelskill.report import Reporter

rep = Reporter(mr)
rep.to_markdown()

Very basic first example report