5 examples of 'select specific columns in pandas' in Python

197
def select_column(df, token):
198
    column_name = token.value
199
    if "." in column_name:
200
        table_name, column_name = column_name.split(".", 1)
201
        #TODO: Assert table name is correct
202
    #TODO: Catch KeyError exception and raise error saying column does not exist
203
    return df[column_name]

42
def extractCols(df, colnames):
43
    extracted = df[colnames]
44
    df.drop(extracted.columns, axis=1, inplace=True)
45
    return extracted

4119
def __getitem__(self, key):
4120
    if not isinstance(key, tuple):
4121
        key = (key,)
4122
    if len(self._internal._index_map) < len(key):
4123
        raise KeyError("Key length ({}) exceeds index depth ({})"
4124
                       .format(len(key), len(self._internal.index_map)))
4125
4126
    cols = (self._internal.index_scols[len(key):] +
4127
            [self._internal.scol_for(self._internal.column_index[0])])
4128
    rows = [self._internal.scols[level] == index
4129
            for level, index in enumerate(key)]
4130
    sdf = self._internal.sdf \
4131
        .select(cols) \
4132
        .where(reduce(lambda x, y: x & y, rows))
4133
4134
    if len(self._internal._index_map) == len(key):
4135
        # if sdf has one column and one data, return data only without frame
4136
        pdf = sdf.limit(2).toPandas()
4137
        length = len(pdf)
4138
        if length == 1:
4139
            return pdf[self.name].iloc[0]
4140
4141
        key_string = name_like_string(key)
4142
        sdf = sdf.withColumn(SPARK_INDEX_NAME_FORMAT(0), F.lit(key_string))
4143
        internal = _InternalFrame(sdf=sdf, index_map=[(SPARK_INDEX_NAME_FORMAT(0), None)])
4144
        return _col(DataFrame(internal))
4145
4146
    internal = self._internal.copy(
4147
        sdf=sdf,
4148
        index_map=self._internal._index_map[len(key):])
4149
4150
    return _col(DataFrame(internal))

1006
def select_dtypes(self, include=None, exclude=None):
1007
    start = time.time()
1008
    init = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
1009
1010
    _select_dtypes = self._data.select_dtypes(include, exclude)
1011
1012
    finish = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
1013
    self._last_operation_dict['time'] = time.time() - start
1014
    self._last_operation_dict['name'] = 'select_dtypes'
1015
    self._last_operation_dict['mem_usage'] = finish - init
1016
    return _select_dtypes

50
def get_continuous_columns(X):
51
    """Get all continuous features from a pandas DataFrame.
52
53
    This function selects all numeric columns from a pandas
54
    DataFrame that are within the ``float`` family.
55
56
    Parameters
57
    ----------
58
    X : pd.DataFrame
59
        The input dataframe.
60
    """
61
    return X.select_dtypes(include=[float])