ó ‡ˆ\c @s5dZddlmZmZmZddlmZmZmZm Z ddl Z ddl m Z ddl ZddlmZmZmZdd lmZdd lmZdd lmZdd lmZdd lmZe jeƒZeddddddƒZeddddddƒZeddddddƒeddddddƒeddddddƒfZ ed ƒd!„ƒZ!de#e#d"„Z$d#„Z%de&dd$d%„Z'de#d&d$e&e(d'd(ƒe(d)d*ƒfe#e&d+„Z)de&dd,„Z*d-de#d&e&e(d'd(ƒe(d)d*ƒfe#d.„Z+dS(/sÞLabeled Faces in the Wild (LFW) dataset This dataset is a collection of JPEG pictures of famous people collected over the internet, all details are available on the official website: http://vis-www.cs.umass.edu/lfw/ iÿÿÿÿ(tlistdirtmakedirstremove(tdirnametjointexiststisdirN(t LooseVersioni(t get_data_homet _fetch_remotetRemoteFileMetadatai(t deprecated(tBunch(tMemory(tb(t_joblibtfilenameslfw.tgzturls.https://ndownloader.figshare.com/files/5976018tchecksumt@055f7d9c632d7370e6fb4afc7468d40f970c34a80d4c6f50ffec63f5a8d536c0slfw-funneled.tgzs.https://ndownloader.figshare.com/files/5976015t@b47c8422c8cded889dc5a13418c4bc2abbda121092b3533a83306f90d900100aspairsDevTrain.txts.https://ndownloader.figshare.com/files/5976012t@1d454dada7dfeca0e7eab6f65dc4e97a6312d44cf142207be28d688be92aabfaspairsDevTest.txts.https://ndownloader.figshare.com/files/5976009t@7cb06600ea8b2814ac26e946201cdb304296262aad67d046a16a7ec85d0ff87cs pairs.txts.https://ndownloader.figshare.com/files/5976006t@ea42330c62c92989f9d7c03237ed5d591365e89b3e649747777b70e692dc1592sIThis function was deprecated in version 0.20 and will be removed in 0.22.cCs$||jƒ}||jƒ:}|S(s=Scale back to 0-1 range in case of normalization for plotting. .. deprecated:: 0.20 This function was deprecated in version 0.20 and will be removed in 0.22. Parameters ---------- face : array_like The array to scale Returns ------- array_like The scaled array (tmintmax(tfacetscaled((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pyt scale_faceEsc Cstd|ƒ}t|dƒ}t|ƒs7t|ƒnxktD]c}t||jƒ}t|ƒs>|rŽtjd|jƒt |d|ƒq¡t d|ƒ‚q>q>W|rÃt|dƒ}t }nt|dƒ}t }t|ƒs†t||jƒ}t|ƒsA|r.tjd|jƒt |d|ƒqAt d|ƒ‚nd d l } tjd |ƒ| j|d ƒjd |ƒt|ƒn||fS(s0Helper function to download any missing LFW datat data_hometlfw_homesDownloading LFW metadata: %sRs %s is missingt lfw_funneledtlfws!Downloading LFW data (~200MB): %siÿÿÿÿNs$Decompressing the data archive to %ssr:gztpath(RRRRtTARGETSRtloggertinfoRR tIOErrortFUNNELED_ARCHIVEtARCHIVEttarfiletdebugtopent extractallR( Rtfunneledtdownload_if_missingRttargetttarget_filepathtdata_folder_pathtarchivet archive_pathR(((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pyt_check_fetch_lfwbs:           cCs+ddlm}m}tddƒtddƒf}|dkrI|}ntd„t||ƒDƒƒ}|\}}|j|j|j pd} |j|j|j pªd} |dk rét |ƒ}t || ƒ} t || ƒ} nt |ƒ} |st j| | | fdt jƒ} n$t j| | | dfdt jƒ} xát|ƒD]Ó\} }| d dkr†tjd | d| ƒn||ƒ}|jdkr´td |ƒ‚nt j||dt jƒ}|d :}|dk rø|||ƒ}n|s|jd dƒ}n|| | df˜sitdtypeiièsLoading face #%05d / %05dsLFailed to read the image file %s, Please make sure that libjpeg is installedgào@taxis.N(texternals._pilutilR4R5tslicetNonettupletziptstoptstarttsteptfloattinttlentnptzerostfloat32t enumerateR#R)tndimt RuntimeErrortasarraytmean(t file_pathstslice_tcolortresizeR4R5t default_sliceth_slicetw_slicethtwtn_facestfacestit file_pathtimgR((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pyt _load_imgss>      $$    icCsrgg}}x·tt|ƒƒD]£}t||ƒ}t|ƒsGq ngtt|ƒƒD]} t|| ƒ^qZ} t| ƒ} | |kr |jddƒ}|j|g| ƒ|j| ƒq q Wt|ƒ} | dkròtd|ƒ‚ntj |ƒ} tj | |ƒ}t ||||ƒ}tj | ƒ}tj jdƒj|ƒ||||}}||| fS(s~Perform the actual data loading for the lfw people dataset This operation is meant to be cached by a joblib wrapper. t_t is*min_faces_per_person=%d is too restrictivei*(tsortedRRRREtreplacetextendt ValueErrorRFtuniquet searchsortedR\tarangetrandomt RandomStatetshuffle(R0RORPRQtmin_faces_per_persont person_namesRNt person_namet folder_pathtftpathst n_picturesRWt target_namesR.RXtindices((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pyt_fetch_lfw_peopleÊs,  .     gà?iFiÃiNi¬c CsGtd|d|d|ƒ\}} tjd|ƒttjƒtdƒkrjtd|ddd d ƒ} ntd |ddd d ƒ} | jtƒ} | | d |d |d|d|ƒ\} } }| j t | ƒdƒ}t t ƒ}t t|ddƒƒ}|jƒ}WdQX|r"|| fStd|d| d| d|d|ƒS(sr Load the Labeled Faces in the Wild (LFW) people dataset (classification). Download it if necessary. ================= ======================= Classes 5749 Samples total 13233 Dimensionality 5828 Features real, between 0 and 255 ================= ======================= Read more in the :ref:`User Guide `. Parameters ---------- data_home : optional, default: None Specify another download and cache folder for the datasets. By default all scikit-learn data is stored in '~/scikit_learn_data' subfolders. funneled : boolean, optional, default: True Download and use the funneled variant of the dataset. resize : float, optional, default 0.5 Ratio used to resize the each face picture. min_faces_per_person : int, optional, default None The extracted dataset will only retain pictures of people that have at least `min_faces_per_person` different pictures. color : boolean, optional, default False Keep the 3 RGB channels instead of averaging them to a single gray level channel. If color is True the shape of the data has one more dimension than the shape with color = False. slice_ : optional Provide a custom 2D slice (height, width) to extract the 'interesting' part of the jpeg files and avoid use statistical correlation from the background download_if_missing : optional, True by default If False, raise a IOError if the data is not locally available instead of trying to download the data from the source site. return_X_y : boolean, default=False. If True, returns ``(dataset.data, dataset.target)`` instead of a Bunch object. See below for more information about the `dataset.data` and `dataset.target` object. .. versionadded:: 0.20 Returns ------- dataset : dict-like object with the following attributes: dataset.data : numpy array of shape (13233, 2914) Each row corresponds to a ravelled face image of original size 62 x 47 pixels. Changing the ``slice_`` or resize parameters will change the shape of the output. dataset.images : numpy array of shape (13233, 62, 47) Each row is a face image corresponding to one of the 5749 people in the dataset. Changing the ``slice_`` or resize parameters will change the shape of the output. dataset.target : numpy array of shape (13233,) Labels associated to each face image. Those labels range from 0-5748 and correspond to the person IDs. dataset.DESCR : string Description of the Labeled Faces in the Wild (LFW) dataset. (data, target) : tuple if ``return_X_y`` is True .. versionadded:: 0.20 RR,R-s Loading LFW people faces from %ss0.12tcachedirtcompressitverboseitlocationRQRiRPROiÿÿÿÿtdescrslfw.rstNtdatatimagesR.RptDESCR(R3R#R)RRt __version__R tcacheRrtreshapeRERt__file__R*RtreadR (RR,RQRiRPROR-t return_X_yRR0tmt load_funcRXR.RptXt module_pathtrst_filetfdescr((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pytfetch_lfw_peopleós(Q     c Cs¡t|dƒ5}g|D]!}|jƒjtdƒƒ^q}WdQXg|D]}t|ƒdkrM|^qM} t| ƒ} tj| dtjƒ} tƒ} x„t | ƒD]v\} }t|ƒdkrd| | <|dt|dƒdf|dt|dƒdff}nut|ƒd krmd| | <|dt|dƒdf|dt|dƒdff}nt d | d|fƒ‚x—t |ƒD]‰\}\}}yt ||ƒ}Wn)t k rät |t |d ƒƒ}nXttt|ƒƒƒ}t |||ƒ}| j|ƒq”Wq«Wt| |||ƒ}t|jƒ}|jdƒ}|jddƒ|jd|dƒ||_|| tjd d gƒfS(s}Perform the actual data loading for the LFW pairs dataset This operation is meant to be cached by a joblib wrapper. trbs NiR9iiiisinvalid line %d: %rsUTF-8sDifferent personss Same person(R*tstriptsplitRRERFRGRDtlistRIRbRt TypeErrortstrR_RtappendR\tshapetpoptinserttarray(tindex_file_pathR0RORPRQt index_filetlnt split_linestslt pair_specstn_pairsR.RNRYt componentstpairtjtnametidxt person_foldert filenamesRZtpairsRRW((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pyt_fetch_lfw_pairsks>4+   $ $  ttrainc Cs–td|d|d|ƒ\}}tjd||ƒttjƒtdƒkrmtd|ddd d ƒ} ntd |ddd d ƒ} | jtƒ} id d 6dd6dd6} || krét d|t t | j ƒƒƒfƒ‚nt || |ƒ} | | |d|d|d|ƒ\} }}ttƒ}tt |ddƒƒ}|jƒ}WdQXtd| jt| ƒdƒd| d|d|d|ƒS(sˆ Load the Labeled Faces in the Wild (LFW) pairs dataset (classification). Download it if necessary. ================= ======================= Classes 5749 Samples total 13233 Dimensionality 5828 Features real, between 0 and 255 ================= ======================= In the official `README.txt`_ this task is described as the "Restricted" task. As I am not sure as to implement the "Unrestricted" variant correctly, I left it as unsupported for now. .. _`README.txt`: http://vis-www.cs.umass.edu/lfw/README.txt The original images are 250 x 250 pixels, but the default slice and resize arguments reduce them to 62 x 47. Read more in the :ref:`User Guide `. Parameters ---------- subset : optional, default: 'train' Select the dataset to load: 'train' for the development training set, 'test' for the development test set, and '10_folds' for the official evaluation set that is meant to be used with a 10-folds cross validation. data_home : optional, default: None Specify another download and cache folder for the datasets. By default all scikit-learn data is stored in '~/scikit_learn_data' subfolders. funneled : boolean, optional, default: True Download and use the funneled variant of the dataset. resize : float, optional, default 0.5 Ratio used to resize the each face picture. color : boolean, optional, default False Keep the 3 RGB channels instead of averaging them to a single gray level channel. If color is True the shape of the data has one more dimension than the shape with color = False. slice_ : optional Provide a custom 2D slice (height, width) to extract the 'interesting' part of the jpeg files and avoid use statistical correlation from the background download_if_missing : optional, True by default If False, raise a IOError if the data is not locally available instead of trying to download the data from the source site. Returns ------- The data is returned as a Bunch object with the following attributes: data : numpy array of shape (2200, 5828). Shape depends on ``subset``. Each row corresponds to 2 ravel'd face images of original size 62 x 47 pixels. Changing the ``slice_``, ``resize`` or ``subset`` parameters will change the shape of the output. pairs : numpy array of shape (2200, 2, 62, 47). Shape depends on ``subset`` Each row has 2 face images corresponding to same or different person from the dataset containing 5749 people. Changing the ``slice_``, ``resize`` or ``subset`` parameters will change the shape of the output. target : numpy array of shape (2200,). Shape depends on ``subset``. Labels associated to each pair of images. The two label values being different persons or the same person. DESCR : string Description of the Labeled Faces in the Wild (LFW) dataset. RR,R-sLoading %s LFW pairs from %ss0.12RsRtiRuiRvspairsDevTrain.txtR£spairsDevTest.txtttests pairs.txtt10_foldss+subset='%s' is invalid: should be one of %rRQRPRORwslfw.rstNRxiÿÿÿÿR¡R.RpRz(R3R#R)RRR{R R|R¢RbR‹R_tkeysRRR~R*RR R}RE(tsubsetRR,RQRPROR-RR0RR‚tlabel_filenamesR“R¡R.RpR„R…R†((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pytfetch_lfw_pairsžs2Q   % $ (,t__doc__tosRRRtos.pathRRRRtloggingtdistutils.versionRtnumpyRFtbaseRR R tutilsR R t utils._joblibR t externals.sixRRt getLoggert__name__R#R'R&R"RR=tTrueR3R\tFalseRrR<R‡R¢R©(((s3lib/python2.7/site-packages/sklearn/datasets/lfw.pyts\"     + =  ( u2