Skip to content

🔭  Dataframe explorer UI

Submitted by Streamlit Data Team!

Summary

Let your viewers explore dataframes themselves! Learn more about it on this blog post

Functions

dataframe_explorer

Adds a UI on top of a dataframe to let viewers filter columns

Parameters:

Name Type Description Default
df DataFrame

Original dataframe

 required
case bool

If True, text inputs will be case sensitive. Defaults to True.

 True

Returns:

Type Description
DataFrame

pd.DataFrame: Filtered dataframe
Source code in src/streamlit_extras/dataframe_explorer/__init__.py 
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
@extra
def dataframe_explorer(df: pd.DataFrame, case: bool = True) -> pd.DataFrame:
    """
    Adds a UI on top of a dataframe to let viewers filter columns

    Args:
        df (pd.DataFrame): Original dataframe
        case (bool, optional): If True, text inputs will be case sensitive. Defaults to True.

    Returns:
        pd.DataFrame: Filtered dataframe
    """

    random_key_base = pd.util.hash_pandas_object(df)

    df = df.copy()

    # Try to convert datetimes into standard format (datetime, no timezone)
    for col in df.columns:
        if is_object_dtype(df[col]):
            try:
                df[col] = pd.to_datetime(df[col])
            except Exception:
                pass

        if is_datetime64_any_dtype(df[col]):
            df[col] = df[col].dt.tz_localize(None)

    modification_container = st.container()

    with modification_container:
        to_filter_columns = st.multiselect(
            "Filter dataframe on",
            df.columns,
            key=f"{random_key_base}_multiselect",
        )
        filters: Dict[str, Any] = dict()
        for column in to_filter_columns:
            left, right = st.columns((1, 20))
            # Treat columns with < 10 unique values as categorical
            if is_categorical_dtype(df[column]) or df[column].nunique() < 10:
                left.write("↳")
                filters[column] = right.multiselect(
                    f"Values for {column}",
                    df[column].unique(),
                    default=list(df[column].unique()),
                    key=f"{random_key_base}_{column}",
                )
                df = df[df[column].isin(filters[column])]
            elif is_numeric_dtype(df[column]):
                left.write("↳")
                _min = float(df[column].min())
                _max = float(df[column].max())
                step = (_max - _min) / 100
                filters[column] = right.slider(
                    f"Values for {column}",
                    _min,
                    _max,
                    (_min, _max),
                    step=step,
                    key=f"{random_key_base}_{column}",
                )
                df = df[df[column].between(*filters[column])]
            elif is_datetime64_any_dtype(df[column]):
                left.write("↳")
                filters[column] = right.date_input(
                    f"Values for {column}",
                    value=(
                        df[column].min(),
                        df[column].max(),
                    ),
                    key=f"{random_key_base}_{column}",
                )
                if len(filters[column]) == 2:
                    filters[column] = tuple(map(pd.to_datetime, filters[column]))
                    start_date, end_date = filters[column]
                    df = df.loc[df[column].between(start_date, end_date)]
            else:
                left.write("↳")
                filters[column] = right.text_input(
                    f"Pattern in {column}",
                    key=f"{random_key_base}_{column}",
                )
                if filters[column]:
                    df = df[df[column].str.contains(filters[column], case=case)]

    return df

Import:

from streamlit_extras.dataframe_explorer import dataframe_explorer # (1)!
  1. You should add this to the top of your .py file 🛠

Examples

example_one

def example_one():
    dataframe = generate_fake_dataframe(
        size=500, cols="dfc", col_names=("date", "income", "person"), seed=1
    )
    filtered_df = dataframe_explorer(dataframe, case=False)
    st.dataframe(filtered_df, use_container_width=True)
Output (beta)