Command Line Examples

Below are examples of how to run modules or example scripts from the command line. Each module/script can take a path to a .json file, providing a configuration, as argument. Most of the examples below do not take in such an argument, and as such will rely on default configurations, which are printed to the screen in purple for reference.

Read in Flat Files, Store in HDF5 File

read data from flat files (.csv, .tsv, etc) and store them in a single .h5

$ python -m GPSat.read_and_store

Plot Raw Observations

e.g. those stored in the above

$ python -m examples.plot_observations

Bin Raw Observations

pre-processing data: in this case remove outliers and bin data.

$ python -m GPSat.bin_data

Run Optimal Interpolation (OI) on Binned Observations

$ python -m examples.local_expert_oi

Post Process OI Results: Smooth Hyper Parameters

run post processing: in this case smooth of hyper parameters

$ python -m GPSat.postprocessing

Run OI by Loading Hyper-Parameters

in this case load the smoothed hyper parameters from above, don’t optimise.

this is done by providing a configuration file, generated in the previous step.

$ python -m examples.local_expert_oi results/example/ABC_binned_example_SMOOTHED.json

Generate Plots of the Predictions

selecting the predictions generated when using the smoothed hyper parameters

$ python -m examples.plot_from_results