View data¶

ParamTools provides 2 primary ways for you to view and access your parameter values:

Using .sel: params.sel["parameter_name"]
As NumPy arrays: params.parameter_name

.sel is for querying parameter values based on their labels or values. Accessing values as NumPy arrays is for using parameter values for computational tasks.

First, let’s create some data.

import paramtools


class Params(paramtools.Parameters):
    defaults = {
        "schema": {
            "labels": {
                "date": {
                    "type": "date", 
                    "validators": {
                        "range": {"min": "2020-01-01", "max": "2021-01-01", "step": {"months": 1}}
                    }
                }
            },
        },
        "a": {
            "title": "A",
            "type": "int",
            "value": [{"date": "2020-01-01", "value": 2}, {"date": "2020-10-01", "value": 8},]
        },
        "b": {
            "title": "B",
            "type": "float",
            "value": [{"date": "2020-01-01", "value": 10.5}]
        }
    }

Access values using `.sel`¶

params = Params()

View all of the values of a:

params.sel["a"]

Values([
  {'date': datetime.date(2020, 1, 1), 'value': 2},
  {'date': datetime.date(2020, 10, 1), 'value': 8},
])

Access parameters directly using isel:

params.sel["a"].isel[0]

{'date': datetime.date(2020, 1, 1), 'value': 2}

Access labels and values directly:

params.sel["a"]["date"][0]

datetime.date(2020, 1, 1)

params.sel["a"]["value"][0]

ParamTools supports slicing:

params.sel["a"]["date"][:2]

[datetime.date(2020, 1, 1), datetime.date(2020, 10, 1)]

Use `label_to_extend` to extend a and b’s values over multiple months.¶

params = Params(label_to_extend="date")
params.sel["a"]

Values([
  {'date': datetime.date(2020, 1, 1), 'value': 2},
  {'date': datetime.date(2020, 2, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 3, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 4, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 5, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 6, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 7, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 8, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 9, 1), 'value': 2, '_auto': True},
  {'date': datetime.date(2020, 10, 1), 'value': 8},
  {'date': datetime.date(2020, 11, 1), 'value': 8, '_auto': True},
  {'date': datetime.date(2020, 12, 1), 'value': 8, '_auto': True},
  {'date': datetime.date(2021, 1, 1), 'value': 8, '_auto': True},
])

Get all values with a date greater than August 1, 2020:

from datetime import date

params.sel["a"]["date"] > date(2020, 8, 1)

QueryResult([
  {'date': datetime.date(2020, 9, 1), 'value': 2, '_auto': True}
  {'date': datetime.date(2020, 10, 1), 'value': 8}
  {'date': datetime.date(2020, 11, 1), 'value': 8, '_auto': True}
  {'date': datetime.date(2020, 12, 1), 'value': 8, '_auto': True}
  {'date': datetime.date(2021, 1, 1), 'value': 8, '_auto': True}
])

Iterate over the query results:

for val in params.sel["a"]["date"] > date(2020, 8, 1):
    print(val)

OrderedDict([('date', datetime.date(2020, 9, 1)), ('value', 2), ('_auto', True)])
OrderedDict([('date', datetime.date(2020, 10, 1)), ('value', 8)])
OrderedDict([('date', datetime.date(2020, 11, 1)), ('value', 8), ('_auto', True)])
OrderedDict([('date', datetime.date(2020, 12, 1)), ('value', 8), ('_auto', True)])
OrderedDict([('date', datetime.date(2021, 1, 1)), ('value', 8), ('_auto', True)])

Convert first two values of the results to a list:

(params.sel["a"]["date"] > date(2020, 8, 1)).isel[:2]

[{'date': datetime.date(2020, 9, 1), 'value': 2, '_auto': True},
 {'date': datetime.date(2020, 10, 1), 'value': 8}]

Access values directly as numpy arrays¶

ParamTools uses information about the parameter labels to convert values from a list of dictionaries to a NumPy array. Set array_first to True for ParamTools to convert parameter values to arrays by default.

params = Params(label_to_extend="date", array_first=True)

params.a

array([2, 2, 2, 2, 2, 2, 2, 2, 2, 8, 8, 8, 8])

params.a * params.b

array([21., 21., 21., 21., 21., 21., 21., 21., 21., 84., 84., 84., 84.])

Select the values of the array:

from datetime import date

arr = params.to_array("a", date=[date(2020, 1, 1), date(2020, 10, 1)])

arr

array([2, 8])

Convert back to a list of dictionaries:

params.from_array("a", arr, date=[date(2020, 1, 1), date(2020, 10, 1)])

[{'date': datetime.date(2020, 1, 1), 'value': 2},
 {'date': datetime.date(2020, 10, 1), 'value': 8}]

Convert parameter values to a dictionary:

param_dict = params.to_dict()

param_dict

{'a': array([2, 2, 2, 2, 2, 2, 2, 2, 2, 8, 8, 8, 8]),
 'b': array([10.5, 10.5, 10.5, 10.5, 10.5, 10.5, 10.5, 10.5, 10.5, 10.5, 10.5,
        10.5, 10.5])}

Convert params to a pandas DataFrame:

import pandas as pd

param_dict["date"] = params.label_grid["date"]

df = pd.DataFrame.from_dict(param_dict)
df.set_index("date")

	a	b
date
2020-01-01	2	10.5
2020-02-01	2	10.5
2020-03-01	2	10.5
2020-04-01	2	10.5
2020-05-01	2	10.5
2020-06-01	2	10.5
2020-07-01	2	10.5
2020-08-01	2	10.5
2020-09-01	2	10.5
2020-10-01	8	10.5
2020-11-01	8	10.5
2020-12-01	8	10.5
2021-01-01	8	10.5

Miscellaneous¶

Get parameter names:

params.keys()

odict_keys(['a', 'b'])

Iterate over parameter values:

for key, value in params.items():
    print(key, value)

a [2 2 2 2 2 2 2 2 2 8 8 8 8]
b [10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5]

Dump all parameter data:

params.dump()

{'schema': {'labels': {'date': {'validators': {'range': {'max': '2021-01-01',
      'min': '2020-01-01',
      'step': {'months': 1}}},
    'number_dims': 0,
    'type': 'date'}},
  'operators': {'array_first': True,
   'uses_extend_func': False,
   'label_to_extend': 'date'},
  'additional_members': {}},
 'a': OrderedDict([('title', 'A'),
              ('type', 'int'),
              ('number_dims', 0),
              ('value',
               [OrderedDict([('date', '2020-01-01'), ('value', 2)]),
                OrderedDict([('date', '2020-02-01'), ('value', 2)]),
                OrderedDict([('date', '2020-03-01'), ('value', 2)]),
                OrderedDict([('date', '2020-04-01'), ('value', 2)]),
                OrderedDict([('date', '2020-05-01'), ('value', 2)]),
                OrderedDict([('date', '2020-06-01'), ('value', 2)]),
                OrderedDict([('date', '2020-07-01'), ('value', 2)]),
                OrderedDict([('date', '2020-08-01'), ('value', 2)]),
                OrderedDict([('date', '2020-09-01'), ('value', 2)]),
                OrderedDict([('date', '2020-10-01'), ('value', 8)]),
                OrderedDict([('date', '2020-11-01'), ('value', 8)]),
                OrderedDict([('date', '2020-12-01'), ('value', 8)]),
                OrderedDict([('date', '2021-01-01'), ('value', 8)])]),
              ('validators', {})]),
 'b': OrderedDict([('title', 'B'),
              ('type', 'float'),
              ('number_dims', 0),
              ('value',
               [OrderedDict([('date', '2020-01-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-02-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-03-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-04-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-05-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-06-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-07-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-08-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-09-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-10-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-11-01'), ('value', 10.5)]),
                OrderedDict([('date', '2020-12-01'), ('value', 10.5)]),
                OrderedDict([('date', '2021-01-01'), ('value', 10.5)])]),
              ('validators', {})])}

Guide Extend

ParamTools

View data¶

Access values using .sel¶

Use label_to_extend to extend a and b’s values over multiple months.¶

Access values directly as numpy arrays¶

Miscellaneous¶

Access values using `.sel`¶

Use `label_to_extend` to extend a and b’s values over multiple months.¶