Na}{

File Content: `mlcomp.cpython-35.opt-1.pyc`



��(X������������������@���s_���d��Z��d�d�l�Z�d�d�l�Z�d�d�l�m�Z�d�d�d���Z�d�e�i�Z�d�d�d�d	���Z�d�S)
zBGlue code to load http://mlcomp.org data as a scikit.learn dataset�����N)�
load_filesc�������������K���s:���|�d��k	�r!�t��j�j�|��|���}��t�|��|�j�d���|���S)N�description)�os�path�joinr����get)�dataset_path�metadata�set_�kwargs��r����
/mlcomp.py�_load_document_classification
���s����r���ZDocumentClassification�rawc�������������K���s|��|�d�k�r>�y�t��j�d�}�Wn�t�k
�r=�t�d�����Yn�Xt��j�j�|���}�t��j�j�|���}�t��j�j�|���}�t��j�j�|���s��t�d�|�����t	�|��t
�j���r��t��j�j�|�t
�|������}�n��d�}�d�|��}�x��t��j�|���D]��}�t��j�j�|�|�d���}�t��j�j�|���sq��t�|����@�}�x6�|�D].�}	�|	�j����|�k�r/t��j�j�|�|���}�Pq/WWd�QRXq��W|�d�k�r�t�d�|�����t����}
�t��j�j�|�d���}�t��j�j�|���s�t�|�d�����t�|����R�}�xH�|�D]@�}	�d	�|	�k�r�|	�j�d	�d
���\�}�}�|�j����|
�|�j����<q�WWd�QRX|
�j�d�d���}
�t�j�|
���}�|�d�k�rft�d
�|
�����|�|�|
�d�|�|��S)a���Load a datasets as downloaded from http://mlcomp.org

    Parameters
    ----------

    name_or_id : the integer id or the string name metadata of the MLComp
                 dataset to load

    set_ : select the portion to load: 'train', 'test' or 'raw'

    mlcomp_root : the filesystem path to the root folder where MLComp datasets
                  are stored, if mlcomp_root is None, the MLCOMP_DATASETS_HOME
                  environment variable is looked up instead.

    **kwargs : domain specific kwargs to be passed to the dataset loader.

    Read more in the :ref:`User Guide <datasets>`.

    Returns
    -------

    data : Bunch
        Dictionary-like object, the interesting attributes are:
        'filenames', the files holding the raw to learn, 'target', the
        classification labels (integer index), 'target_names',
        the meaning of the labels, and 'DESCR', the full description of the
        dataset.

    Note on the lookup process: depending on the type of name_or_id,
    will choose between integer id lookup or metadata name lookup by
    looking at the unzipped archives and metadata file.

    TODO: implement zip dataset loading too
    NZMLCOMP_DATASETS_HOMEz.MLCOMP_DATASETS_HOME env variable is undefinedzCould not find folder: zname: r	���z+Could not find dataset with metadata line: z is not a valid MLComp dataset�:�����formatZunknowz"No loader implemented for format: r
���)r����environ�KeyError�
ValueErrorr����
expanduser�abspath�normpath�exists�
isinstance�numbersZIntegralr����str�listdir�open�strip�dict�splitr����LOADERS)Z
name_or_idr
���Zmlcomp_rootr���r���Zexpected_name_lineZdatasetZ
metadata_file�f�liner	����key�valuer����loaderr���r���r
����load_mlcomp���sP����$



	
!r(���)�__doc__r���r���Zsklearn.datasets.baser���r���r"���r(���r���r���r���r
����<module>���s���

Name	Size	Permission
__init__.cpython-35.opt-1.pyc	3078 bytes	0644
__init__.cpython-35.pyc	3078 bytes	0644
base.cpython-35.opt-1.pyc	24686 bytes	0644
base.cpython-35.pyc	24686 bytes	0644
california_housing.cpython-35.opt-1.pyc	3623 bytes	0644
california_housing.cpython-35.pyc	3623 bytes	0644
covtype.cpython-35.opt-1.pyc	3595 bytes	0644
covtype.cpython-35.pyc	3595 bytes	0644
kddcup99.cpython-35.opt-1.pyc	12034 bytes	0644
kddcup99.cpython-35.pyc	12034 bytes	0644
lfw.cpython-35.opt-1.pyc	16547 bytes	0644
lfw.cpython-35.pyc	16547 bytes	0644
mlcomp.cpython-35.opt-1.pyc	3126 bytes	0644
mlcomp.cpython-35.pyc	3126 bytes	0644
mldata.cpython-35.opt-1.pyc	6339 bytes	0644
mldata.cpython-35.pyc	6339 bytes	0644
olivetti_faces.cpython-35.opt-1.pyc	4514 bytes	0644
olivetti_faces.cpython-35.pyc	4514 bytes	0644
rcv1.cpython-35.opt-1.pyc	7184 bytes	0644
rcv1.cpython-35.pyc	7184 bytes	0644
samples_generator.cpython-35.opt-1.pyc	51364 bytes	0644
samples_generator.cpython-35.pyc	51364 bytes	0644
setup.cpython-35.opt-1.pyc	814 bytes	0644
setup.cpython-35.pyc	814 bytes	0644
species_distributions.cpython-35.opt-1.pyc	7434 bytes	0644
species_distributions.cpython-35.pyc	7434 bytes	0644
svmlight_format.cpython-35.opt-1.pyc	15631 bytes	0644
svmlight_format.cpython-35.pyc	15631 bytes	0644
twenty_newsgroups.cpython-35.opt-1.pyc	12307 bytes	0644
twenty_newsgroups.cpython-35.pyc	12307 bytes	0644

File Content: mlcomp.cpython-35.opt-1.pyc

File Content: `mlcomp.cpython-35.opt-1.pyc`