Usage

As mentioned earlier, the datacube-stats allows the user to worry only about the definition of the operation that will be carried out. The way the process will be done is up to the tool. The description of the operation is made using YAML files. This section will present an example of a YAML file describing a simple NDVI mean operation and how the datacube-stats consume and execute that operation.

To start, the mean operation in the NDVI “band” that is delivered in the BDC’s datacubes is done based on the file presented below

Code snippet 5 - CB4_64_16D_STK_1.temporal_mean.stats.yaml File with the description of the mean operation in NDVI

sources:
  - product: CB4_64_16D_STK_1
    measurements: [NDVI]
    group_by: solar_day
    time: [2020-07-11, 2020-07-27]

date_ranges:
  start_date: 2020-07-01
  end_date:  2020-08-01

location: /data/stats/

storage:
  driver: GeoTIFF

  crs: +proj=aea +lat_0=-12 +lon_0=-54 +lat_1=-2 +lat_2=-22 +x_0=5000000 +y_0=10000000 +ellps=GRS80 +units=m +no_defs
  tile_size:
    x: 100000.0
    y: 100000.0
  resolution:
    x: 64
    y: -64
  chunking:
    x: 256
    y: 256
    time: 1
  dimension_order: [time, x, y]

computation:
  chunking:
    x: 1000
    y: 1000

output_products:
  - name: CB4_64_16D_STK_1_simple_mean
    product_type: ndvi_simple_mean
    statistic: simple
    statistic_args:
      reduction_function: mean
    output_params:
      zlib: True
      fletcher32: True
    metadata:
      format:
        name: GeoTIFF
      platform:
        code: CBERS4
      instrument:
        name: AWFI
    file_path_template: 'CB4_64_16D_STK_1_NDVI_SIMPLE_MEAN/{name}_ndvi_{y}_{x}_{epoch_start:%Y-%m-%d}_{epoch_end:%Y-%m-%d}.tif'

Note that in the file presented above, the form of storage, transformations, and the statistical operation performed is described, but at no time does the user need to describe how the operation will be performed.

Note

To learn more about the operations description file, consult the datacube-stats repository.

After defining the file, it can be consumed using the CLI of datacube-stats. This consumption can be done by passing several parameters, which can change how the operation is performed. For example, to run on a single processor, using 6 cores, the --parallel parameter can be used:

datacube-stats --parallel 6 CB4_64_16D_STK_1.temporal_mean.stats.yaml

o execute with the dask, the operation can be done using the parameter --dask, passing the scheduler’s connection parameters:

datacube-stats --dask scheduler:8786 CB4_64_16D_STK_1.temporal_mean.stats.yaml

Note

To execute the operations above, you must connect to the client container, which was started with the others, use the command below:

docker-compose exec odc-stats bash

Note

For more files with descriptions of operations considering BDC data products, consult the BDC-ODC repository.