Before I forget: somewhere prominently the following should be acknowledged in some form, not necessarily in the below:

Bernd Bischl, who mentioned the idea of a general data container with transformers attached to columns in personal discussion during a London visit in 2016. Myself, having (in my opinion) substantially contributed through the API design (?). Haoran Xue, who completed a thesis on the topic erlier. While no code was transferred, lessons that were learnt may have been transferred.

This pull request improves the readability of the documentation.

While going through your codebase, I realised that there's a lot of redundancy in the module naming, e.g. /transformers/transformers/series_transformers/series_transformer.py instead of /transformers/series/series_transformer.py. Is there any specific reason for that? If not I'd suggest you to refactor the module into a more straightforward naming structure.

currently it adds the transformer output while retaining the original column

for retaining original column: use identityTransformer (to be implemented)

Structure should be changed to: (1) data container (Xseries and XDataFrame) (2) transformer functionality

since user should be made aware that (1) is a separate interface concept on top of which (2) may be invoked but isn't necessarily tied together

Bumps numpy from 1.15.2 to 1.22.0.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Bumps ipython from 7.0.1 to 7.16.3.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

great code thanks may you clarify : will it work for multivariate time series prediction both regression and classification 1 where all values are continues values weight height age target 1 56 160 34 1.2 2 77 170 54 3.5 3 87 167 43 0.7 4 55 198 72 0.5 5 88 176 32 2.3

2 or even will it work for multivariate time series where values are mixture of continues and categorical values for example 2 dimensions have continues values and 3 dimensions are categorical values

color        weight     gender  height  age  target 

1 black 56 m 160 34 yes 2 white 77 f 170 54 no 3 yellow 87 m 167 43 yes 4 white 55 m 198 72 no 5 white 88 f 176 32 yes

great code thanks may you clarify : will it work for multivariate time series prediction both regression and classification 1 where all values are continues values weight height age target 1 56 160 34 1.2 2 77 170 54 3.5 3 87 167 43 0.7 4 55 198 72 0.5 5 88 176 32 2.3

2 or even will it work for multivariate time series where values are mixture of continues and categorical values for example 2 dimensions have continues values and 3 dimensions are categorical values

color        weight     gender  height  age  target 

1 black 56 m 160 34 yes 2 white 77 f 170 54 no 3 yellow 87 m 167 43 yes 4 white 55 m 198 72 no 5 white 88 f 176 32 yes

# loading some time-series data
from io import BytesIO
from zipfile import ZipFile
from urllib.request import urlopen
from xpandas.data_container import XSeries, XDataFrame
import numpy as np
import pandas as pd

def read_data(file):
    data = file.readlines()
    rows = [row.decode('utf-8').strip().split('  ') for row in data]
    X = pd.DataFrame(rows, dtype=np.float)
    y = X.pop(0)
    ts = XSeries([row for _, row in X.iterrows()])
    X = XDataFrame({'ts1': ts, 'ts2': ts})
    return X, y

url = 'http://www.timeseriesclassification.com/Downloads/GunPoint.zip'
url = urlopen(url)
zipfile = ZipFile(BytesIO(url.read()))
file = zipfile.open('GunPoint_TRAIN.txt')
X, y = read_data(file)

X.mean() # returns empty series rather than mean of series, the same for many other methods like .std(), .median(), etc)

X.apply(np.mean) # breaks

X['ts1'].mean() # breaks 

X['ts1'].apply(np.mean) # works

X['ts1'].apply(np.percentile, args=(25,)) # breaks, does not passes on args 

Slicing of single row in XDataFrame does not work, probably because it tries to return a series which does not work as types are heterogeneous, so instead one may want to return a XDataFrame with a single row.

import pandas as pd
from xpandas.data_container import XDataFrame

iris = pd.read_csv('https://raw.githubusercontent.com/mwaskom/seaborn-data/master/iris.csv')
iris.iloc[0] # works

irisx = XDataFrame(iris) 
irisx.iloc[0] # breaks

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-32-b46ce52f9af0> in <module>
----> 1 irisx.iloc[0]

~/.conda/envs/sktime/lib/python3.7/site-packages/pandas/core/indexing.py in __getitem__(self, key)
   1476 
   1477             maybe_callable = com._apply_if_callable(key, self.obj)
-> 1478             return self._getitem_axis(maybe_callable, axis=axis)
   1479 
   1480     def _is_scalar_access(self, key):

~/.conda/envs/sktime/lib/python3.7/site-packages/pandas/core/indexing.py in _getitem_axis(self, key, axis)
   2102             self._validate_integer(key, axis)
   2103 
-> 2104             return self._get_loc(key, axis=axis)
   2105 
   2106     def _convert_to_indexer(self, obj, axis=None, is_setter=False):

~/.conda/envs/sktime/lib/python3.7/site-packages/pandas/core/indexing.py in _get_loc(self, key, axis)
    143         if axis is None:
    144             axis = self.axis
--> 145         return self.obj._ixs(key, axis=axis)
    146 
    147     def _slice(self, obj, axis=None, kind=None):

~/.conda/envs/sktime/lib/python3.7/site-packages/pandas/core/frame.py in _ixs(self, i, axis)
   2624                                                       index=self.columns,
   2625                                                       name=self.index[i],
-> 2626                                                       dtype=new_values.dtype)
   2627                 result._set_is_copy(self, copy=copy)
   2628                 return result

~/.conda/envs/sktime/lib/python3.7/site-packages/xpandas/data_container/data_container.py in __init__(self, *args, **kwargs)
     71         check_result, data_type = _check_all_elements_have_the_same_property(data, type)
     72         if not check_result:
---> 73             raise ValueError('Not all elements the same type')
     74 
     75         if data_type is not None:

ValueError: Not all elements the same type