Training datasets

datasets.training

datasets.training

This module exposes the datasets to train ocean forecasting systems. They cover the years 1993 to 2019, and never the year evaluated by the benchmark.

Functions

Name	Description
glorys_reanalysis	Open the 1/12 degree GLORYS reanalysis from 1993 to 2019 (included) as an `xarray.Dataset`.
glorys_reanalysis_1_degree	Open the 1 degree GLORYS reanalysis from 1993 to 2019 (included) as an `xarray.Dataset`.

glorys_reanalysis

datasets.training.glorys_reanalysis()

Open the 1/12 degree GLORYS reanalysis from 1993 to 2019 (included) as an xarray.Dataset.

It is opened lazily with the Copernicus Marine Toolbox, and therefore requires authentication with the Copernicus Marine Service.

Returns

Name	Type	Description
	Dataset	The Dataset containing the GLORYS reanalysis from 1993-01-01 to 2019-12-31.

Examples

>>> glorys_reanalysis()
<xarray.Dataset> Size: 140TB
Dimensions:    (depth: 50, latitude: 2041, longitude: 4320, time: 9861)
Coordinates:
  * depth      (depth) float32 200B 0.494 1.541 2.646 ... 5.275e+03 5.728e+03
  * latitude   (latitude) float32 8kB -80.0 -79.92 -79.83 ... 89.83 89.92 90.0
  * longitude  (longitude) float32 17kB -180.0 -179.9 -179.8 ... 179.8 179.9
  * time       (time) datetime64[ns] 79kB 1993-01-01 1993-01-02 ... 2019-12-31
Data variables:
    thetao     (time, depth, latitude, longitude) float64 35TB dask.array<chunksize=(50, 1, 512, 2048), meta=np.ndarray>
    so         (time, depth, latitude, longitude) float64 35TB dask.array<chunksize=(50, 1, 512, 2048), meta=np.ndarray>
    uo         (time, depth, latitude, longitude) float64 35TB dask.array<chunksize=(50, 1, 512, 2048), meta=np.ndarray>
    vo         (time, depth, latitude, longitude) float64 35TB dask.array<chunksize=(50, 1, 512, 2048), meta=np.ndarray>
    zos        (time, latitude, longitude) float64 696GB dask.array<chunksize=(50, 512, 2048), meta=np.ndarray>
Attributes:
    ...

glorys_reanalysis_1_degree

datasets.training.glorys_reanalysis_1_degree()

Open the 1 degree GLORYS reanalysis from 1993 to 2019 (included) as an xarray.Dataset.

This dataset is intended to support training on the 1 degree OceanBench track. It exposes the historical GLORYS reanalysis published at 1 degree resolution in the OceanBench public EDITO bucket.

Returns

Name	Type	Description
	Dataset	The Dataset containing the 1 degree GLORYS reanalysis from 1993-01-01 to 2019-12-31.

Examples

>>> glorys_reanalysis_1_degree()
<xarray.Dataset> Size: 970GB
Dimensions:                          (time: 9861, depth: 50, latitude: 170,
                                      longitude: 360)
Coordinates:
  * depth                            (depth) float32 200B 0.494 ... 5.728e+03
  * latitude                         (latitude) float64 1kB -79.5 -78.5 ... 89.5
  * longitude                        (longitude) float64 3kB -179.5 ... 179.5
  * time                             (time) datetime64[ns] 79kB 1993-01-01 .....
Data variables:
    eastward_sea_water_velocity      (time, depth, latitude, longitude) float64 241GB dask.array<chunksize=(1, 1, 170, 360), meta=np.ndarray>
    northward_sea_water_velocity     (time, depth, latitude, longitude) float64 241GB dask.array<chunksize=(1, 1, 170, 360), meta=np.ndarray>
    sea_surface_height_above_geoid   (time, latitude, longitude) float64 5GB dask.array<chunksize=(1, 170, 360), meta=np.ndarray>
    sea_water_potential_temperature  (time, depth, latitude, longitude) float64 241GB dask.array<chunksize=(1, 1, 170, 360), meta=np.ndarray>
    sea_water_salinity               (time, depth, latitude, longitude) float64 241GB dask.array<chunksize=(1, 1, 170, 360), meta=np.ndarray>
Attributes:
    ...