3 examples of 'how to convert a pandas column to timedelta' in Python

207
def decode_cf_timedelta(num_timedeltas, units):
208
    """Given an array of numeric timedeltas in netCDF format, convert it into a
209
    numpy timedelta64[ns] array.
210
    """
211
    num_timedeltas = np.asarray(num_timedeltas)
212
    units = _netcdf_to_numpy_timeunit(units)
213
    result = to_timedelta_unboxed(num_timedeltas.ravel(), unit=units)
214
    return result.reshape(num_timedeltas.shape)

30
def converter(df: DataFrame) -> Series:
31
    return df[field_name]

107
def _time_cols_to_ints(df, columns=None, nat_value=SMALLDT64):
108
    """
109
    Converts all datetime columns to ints.
110
111
    Previously NaT were convertible to ints, but now they raise a value error.
112
    We need, therefore, to designate a time that will be used as NaT.
113
    """
114
    cols = columns or df.select_dtypes(include=["datetime64"]).columns
115
    df.loc[:, cols] = df.loc[:, cols].fillna(nat_value).astype(np.int64)
116
    return df