python - Create dictionary for classification dataset storage similar to digits dataset (sklearn) -

i extracting features images , save them alongside labels (and original images preferably) able load them later without running code feature extraction every time.

i store them in in similar structure 1 digits dataset in sklearn.datasets, dictionary type.

so problem not storing of type key:value, of type:

• features (x)
• target_labels (y)
• images (optional)
• target_names

my x numpy.ndarray data type , y 1-d vector array.

any suggestions how achieve this?

if want sklearn.datasets methods return why don't use code?

they define class bunch want:

class bunch(dict):     """container object datasets     dictionary-like object exposes keys attributes.     >>> b = bunch(a=1, b=2)     >>> b['b']     2     >>> b.b     2     >>> b.a = 3     >>> b['a']     3     >>> b.c = 6     >>> b['c']     6     """      def __init__(self, **kwargs):         super(bunch, self).__init__(kwargs)      def __setattr__(self, key, value):         self[key] = value      def __dir__(self):         return self.keys()      def __getattr__(self, key):         try:             return self[key]         except keyerror:             raise attributeerror(key) 

and create dataset object with:

bunch(data=data, target=target,                  target_names=target_names,                  descr=fdescr,                  feature_names=['feat_1', 'feat_2',                                 'feat_3', 'feat_4'])