Skip to content

Datasets

Example datasets for latent_calendar.

All datasets are loaded from the web and cached locally if desired with the local_save argument.

The datasets all include one or more columns that represent a datetime that can be used for calendar analysis.

Examples:

Load the chicago bikes dataset:

from latent_calendar.datasets import load_chicago_bikes

df = load_chicago_bikes()
df.head()

                start_station_name end_station_name  rideable_type          started_at            ended_at member_casual
ride_id
ABF4F851DE485B76                NaN              NaN  electric_bike 2023-06-30 18:56:13 2023-06-30 19:30:40        member
F123B5D34B002CDB                NaN              NaN  electric_bike 2023-06-30 06:12:31 2023-06-30 06:23:05        member
CA8E2C38AF641DFB                NaN              NaN  electric_bike 2023-06-30 08:28:51 2023-06-30 08:37:45        member
93CCE4EA48CFDB69                NaN              NaN  electric_bike 2023-06-30 09:09:24 2023-06-30 09:17:41        member
FDBCEFE7890F7262                NaN              NaN  electric_bike 2023-06-30 16:29:48 2023-06-30 16:38:51        member

load_chicago_bikes(local_save=False, **read_kwargs)

Bikesharing trip level data from Chicago's Divvy system.

Read more about the data source here.

The data is two weeks of trips starting June 26th, 2023 until July 9th, 2023.

Parameters:

Name Type Description Default
local_save bool

Whether to save the data locally if it doesn't exists.

 False
read_kwargs

kwargs to pass to pd.read_csv

 {}

Returns:

Type Description
DataFrame

Trips data from Chicago's Divvy system.
Source code in latent_calendar/datasets/__init__.py 
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
def load_chicago_bikes(local_save: bool = False, **read_kwargs) -> pd.DataFrame:
    """Bikesharing trip level data from Chicago's Divvy system.

    Read more about the data source [here](https://data.cityofchicago.org/Transportation/Divvy-Trips/fg6s-gzvg).

    The data is two weeks of trips starting June 26th, 2023 until July 9th, 2023.

    Args:
        local_save: Whether to save the data locally if it doesn't exists.
        read_kwargs: kwargs to pass to pd.read_csv

    Returns:
        Trips data from Chicago's Divvy system.

    """
    name = "chicago-bikes"
    read_kwargs = {
        "parse_dates": ["started_at", "ended_at"],
        "index_col": ["ride_id"],
        **read_kwargs,
    }

    return _load_data(name, read_kwargs=read_kwargs, local_save=local_save)

load_online_transactions(local_save=False, **read_kwargs)

Kaggle Data for an non-store online retailer in UK. More information here.

Parameters:

Name Type Description Default
local_save bool

Whether to save the data locally if it doesn't exists.

 False
read_kwargs

kwargs to pass to pd.read_csv

 {}

Returns:

Type Description
DataFrame

Online transactions data from a non-store online retailer in UK.
Source code in latent_calendar/datasets/__init__.py 
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
def load_online_transactions(local_save: bool = False, **read_kwargs) -> pd.DataFrame:
    """Kaggle Data for an non-store online retailer in UK. More information [here](https://www.kaggle.com/datasets/mashlyn/online-retail-ii-uci).

    Args:
        local_save: Whether to save the data locally if it doesn't exists.
        read_kwargs: kwargs to pass to pd.read_csv

    Returns:
        Online transactions data from a non-store online retailer in UK.

    """
    name = "online_retail_II"
    read_kwargs = {
        "parse_dates": ["InvoiceDate"],
        **read_kwargs,
    }

    return _load_data(name, read_kwargs=read_kwargs, local_save=local_save)

load_ufo_sightings(local_save=False, **read_kwargs)

UFO sightings over time around the world. More info here.

Parameters:

Name Type Description Default
local_save bool

Whether to save the data locally if it doesn't exists.

 False
read_kwargs

kwargs to pass to pd.read_csv

 {}

Returns:

Type Description
DataFrame

Sighting level data for UFOs.
Source code in latent_calendar/datasets/__init__.py 
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
def load_ufo_sightings(local_save: bool = False, **read_kwargs) -> pd.DataFrame:
    """UFO sightings over time around the world. More info [here](https://www.kaggle.com/datasets/camnugent/ufo-sightings-around-the-world).

    Args:
        local_save: Whether to save the data locally if it doesn't exists.
        read_kwargs: kwargs to pass to pd.read_csv

    Returns:
        Sighting level data for UFOs.

    """
    name = "ufo_sighting_data"
    read_kwargs = {"low_memory": False, **read_kwargs}
    save_kwargs = {"index": False}

    df = _load_data(
        name, read_kwargs=read_kwargs, save_kwargs=save_kwargs, local_save=local_save
    )
    df["Date_time"] = pd.to_datetime(
        df["Date_time"]
        .str.replace(" 24:00", " 23:59")
        .pipe(pd.to_datetime, format="mixed", dayfirst=True)
    )

    return df

Comments