Skip to content

Latest commit

 

History

History
106 lines (80 loc) · 5.57 KB

init.md

File metadata and controls

106 lines (80 loc) · 5.57 KB

Creating a BeliefsDataFrame

Table of contents

  1. Sensors and BeliefSources
  2. From a Pandas Series
  3. From a Pandas DataFrame
  4. From a CSV file
  5. From a list of TimedBeliefs

Sensors and BeliefSources

Each belief requires a Sensor and a BeliefSource. For example:

>>> import timely_beliefs as tb
>>> sensor = tb.Sensor("EPEX SPOT day-ahead price", event_resolution=timedelta(hours=1), unit="EUR/MWh")
>>> source = tb.BeliefSource("EPEX")

When creating a BeliefsDataFrame, a sensor always needs to be passed. A BeliefsDataFrame can only refer to a single sensor. A source can be passed once for the entire frame, or per belief.

From a Pandas Series

Create a BeliefsDataFrame from a conventional time series (a Pandas Series with a DatetimeIndex) together with keyword arguments.

Required arguments:

  • either:

    • belief_time (one moment at which all beliefs were formed), or
    • belief_horizon (each belief was formed the same duration before or after)

  • source (one source for all beliefs)

  • sensor (one sensor to which all beliefs refer)

    import pandas as pd import pytz s = pd.Series([63, 60], index=pd.date_range(datetime(2000, 1, 3, 9), periods=2, tz=pytz.utc)) bdf = tb.BeliefsDataFrame(s, belief_horizon=timedelta(hours=0), source=source, sensor=sensor) print(bdf) event_value event_start belief_horizon source cumulative_probability
    2000-01-03 09:00:00+00:00 0 days EPEX 0.5 63 2000-01-04 09:00:00+00:00 0 days EPEX 0.5 60 sensor: <Sensor: EPEX SPOT day-ahead price>, event_resolution: 1:00:00

From a Pandas DataFrame

Pass a Pandas DataFrame with columns ["event_start", "belief_time", "source", "cumulative_probability", "event_value"]. The "cumulative_probability" column is optional (a default of 0.5 will be used if the column is missing).

>>> df = pd.DataFrame([[63, datetime(2000, 1, 3, 9, tzinfo=pytz.utc), timedelta(hours=0), source], [60, datetime(2000, 1, 3, 10, tzinfo=pytz.utc), timedelta(hours=0), source]], columns=["event_value", "event_start", "belief_horizon", "source"])
>>> bdf = tb.BeliefsDataFrame(df, sensor=sensor)
>>> print(bdf)
                                                                        event_value
event_start               belief_horizon source cumulative_probability             
2000-01-03 09:00:00+00:00 0 days         EPEX   0.5                              63
2000-01-03 10:00:00+00:00 0 days         EPEX   0.5                              60
sensor: <Sensor: EPEX SPOT day-ahead price>, event_resolution: 1:00:00

Alternatively, a keyword argument can be used to replace a column that contains the same value for each belief.

>>> df = pd.DataFrame([[63, datetime(2000, 1, 3, 9, tzinfo=pytz.utc)], [60, datetime(2000, 1, 3, 10, tzinfo=pytz.utc)]], columns=["event_value", "event_start"])
>>> bdf = tb.BeliefsDataFrame(df, belief_horizon=timedelta(hours=0), source=source, sensor=sensor)
>>> print(bdf)
                                                                        event_value
event_start               belief_horizon source cumulative_probability             
2000-01-03 09:00:00+00:00 0 days         EPEX   0.5                              63
2000-01-03 10:00:00+00:00 0 days         EPEX   0.5                              60
sensor: <Sensor: EPEX SPOT day-ahead price>, event_resolution: 1:00:00

From a CSV file

The utility function tb.read_csv lets you load a BeliefsDataFrame from a csv file. You still need to set the source and sensor for the BeliefsDataFrame; the csv file only contains their names.

An example, adapted from timely_beliefs/example/__init__.py:

temperature_df = tb.read_csv(
    "temperature.csv",
    sensor=tb.Sensor("Thermometer A", unit="°C", event_resolution=timedelta(hours=1)),
    source=tb.BeliefSource("Source X"),
)

In case the csv file contains multiple source names, you can pass a list of sources. Each source name will then be replaced by the actual Source object from the list you provided. To write a BeliefsDataFrame to a csv file, just use the pandas way:

>>> bdf.to_csv("data.csv")
>>> tb.read_csv("data.csv", source=source, sensor=sensor)

From a list of TimedBeliefs

Create a list of TimedBelief or DBTimedBelief objects and use it to initialize a BeliefsDataFrame.

>>> from datetime import datetime, timedelta
>>> import pytz
>>> belief_1 = tb.TimedBelief(event_value=63, event_start=datetime(2000, 1, 3, 9, tzinfo=pytz.utc), belief_horizon=timedelta(hours=0), sensor=sensor, source=source)
>>> belief_2 = tb.TimedBelief(event_value=60, event_start=datetime(2000, 1, 3, 10, tzinfo=pytz.utc), belief_horizon=timedelta(hours=0), sensor=sensor, source=source)
>>> beliefs = [belief_1, belief_2]
>>> bdf = tb.BeliefsDataFrame(beliefs)
>>> print(bdf)
                                                                                   event_value
event_start               belief_time               source cumulative_probability             
2000-01-03 09:00:00+00:00 2000-01-03 10:00:00+00:00 EPEX   0.5                              63
2000-01-03 10:00:00+00:00 2000-01-03 11:00:00+00:00 EPEX   0.5                              60
sensor: <Sensor: EPEX SPOT day-ahead price>, event_resolution: 1:00:00