pastastore

This module stores Pastas time series and models in a database.

Storing time series and models in a database allows the user to manage time series and Pastas models on disk, which allows the user to pick up where they left off without having to reload everything.

Installation

Install the module with pip install pastastore.

For development, clone the repository and install all development, testing, and documentation dependencies with:

pip install -e .[dev]

For plotting background maps, the contextily and pyproj packages are required. For a full install, including optional dependencies for plotting and labeling data on maps, use: pip install pastastore[optional] Windows users are asked to install rasterio themselves since it often cannot be installed using pip. rasterio is a dependency of contextily.

Usage

The following snippets show typical usage. The first step is to define a so-called Connector object. This object contains methods to store time series or models to the database, or read objects from the database.

The following code creates a PasConnector, which uses Pastas JSON-styled ".pas-files" to save models in a folder on your computer (in this case a folder called pastas_db in the current directory).

import pastastore as pst
create connector instance
conn = pst.PasConnector(name="pastas_db", path=".")

The next step is to pass that connector to the PastaStore object. This object contains all kinds of useful methods to analyze and visualize time series, and build and analyze models.

# create PastaStore instance
pstore = pst.PastaStore(conn)

Now the user can add time series, models or analyze or visualize existing objects in the database. Some examples showing the functionality of the PastaStore object are shown below:

import pandas as pd
import pastas as ps
load oseries from CSV and add to database
oseries = pd.read_csv("oseries.csv")
pstore.addoseries(oseries, "myoseries", metadata={"x": 100000, "y": 400000})
read oseries from database
oseries = pstore.getoseries("myoseries")
view oseries metadata DataFrame
pstore.oseries
plot oseries location on map
ax = pstore.maps.oseries()
pstore.maps.addbackgroundmap(ax)  # add a background map
plot my_oseries time series
ax2 = pstore.plot.oseries(names=["my_oseries"])
create a model with pastas
ml = ps.Model(oseries, name="my_model")
add model to database
pstore.add_model(ml)
load model from database
ml2 = pstore.getmodels("mymodel")
export whole database to a zip file
pstore.tozip("mybackup.zip")

For more elaborate examples, refer to the Notebooks.