list-of-dicts: overview

The list-of-dicts source reads data from a list of dictionaries. Each dictionary represents a single field and the result is a FieldList consisting of ArrayField fields.

First, we create the input data using generic metadata keys. The latitudes and longitudes specified as disctinct values, the 2D field geometry will be automatically built from them.

[1]:

import earthkit.data as ekd

prototype = {
        "latitudes": [10.0, 0.0, -10.0],
        "longitudes": [20, 40.0],
        "values": [1, 2, 3, 4, 5, 6],
        "valid_datetime": "2018-08-01T09:00:00Z",
    }

d = [
        {"param": "t", "level": 500, **prototype},
        {"param": "t", "level": 850, **prototype},
        {"param": "u", "level": 500, **prototype},
        {"param": "u", "level": 850, **prototype},
        {"param": "d", "level": 850, **prototype},
        {"param": "d", "level": 600, **prototype},
    ]

Next, load the data into fieldlist as a list-of-dicts source using from_source().

[2]:

ds = ekd.from_source("list-of-dicts", d)

[3]:

ds.ls()

[3]:

	param	level	base_datetime	valid_datetime	step	number
0	t	500	None	2018-08-01T09:00:00+00:00	None	None
1	t	850	None	2018-08-01T09:00:00+00:00	None	None
2	u	500	None	2018-08-01T09:00:00+00:00	None	None
3	u	850	None	2018-08-01T09:00:00+00:00	None	None
4	d	850	None	2018-08-01T09:00:00+00:00	None	None
5	d	600	None	2018-08-01T09:00:00+00:00	None	None

[4]:

ds[0].to_latlon()

[4]:

{'lat': array([[ 10.,  10.],
        [  0.,   0.],
        [-10., -10.]]),
 'lon': array([[20., 40.],
        [20., 40.],
        [20., 40.]])}

[5]:

ds[0].grid_points()

[5]:

(array([ 10.,  10.,   0.,   0., -10., -10.]),
 array([20., 40., 20., 40., 20., 40.]))

[6]:

ds.to_numpy()

[6]:

array([[[1, 2],
        [3, 4],
        [5, 6]],

       [[1, 2],
        [3, 4],
        [5, 6]],

       [[1, 2],
        [3, 4],
        [5, 6]],

       [[1, 2],
        [3, 4],
        [5, 6]],

       [[1, 2],
        [3, 4],
        [5, 6]],

       [[1, 2],
        [3, 4],
        [5, 6]]])

[7]:

ds.metadata("level")

[7]:

[500, 850, 500, 850, 850, 600]

[8]:

ds[0].metadata().datetime()

[8]:

{'base_time': None,
 'valid_time': datetime.datetime(2018, 8, 1, 9, 0, tzinfo=tzutc())}

[9]:

ds[0].metadata().valid_datetime()

[9]:

datetime.datetime(2018, 8, 1, 9, 0, tzinfo=tzutc())

[10]:

ds.sel(param="t").ls()

[10]:

	param	level	base_datetime	valid_datetime	step	number
0	t	500	None	2018-08-01T09:00:00+00:00	None	None
1	t	850	None	2018-08-01T09:00:00+00:00	None	None

[11]:

ds.order_by("level").ls()

[11]:

	param	level	base_datetime	valid_datetime	step	number
0	t	500	None	2018-08-01T09:00:00+00:00	None	None
1	u	500	None	2018-08-01T09:00:00+00:00	None	None
2	d	600	None	2018-08-01T09:00:00+00:00	None	None
3	t	850	None	2018-08-01T09:00:00+00:00	None	None
4	u	850	None	2018-08-01T09:00:00+00:00	None	None
5	d	850	None	2018-08-01T09:00:00+00:00	None	None