=h)$SSKJr SSKJr SSKJrJr SSKJrJ r J r J r J r SSK r SSKJrJr SSKJr SSKJr SSKrSS KJrJrJr SS jr"S S \5rSS jrSrSr Sr!Sr"Sr#SSjr$Sr%SSjr&SSjr'SSjr(SSjr)Sr*S Sjr+g)!)lrange)StringIO)environmakedirs)abspathdirnameexists expanduserjoinN) HTTPErrorURLError)urljoin)urlopen)Indexread_csv read_statac4[XS-5n[U5$)a Download and return an example dataset from Stata. Parameters ---------- data : str Name of dataset to fetch. baseurl : str The base URL to the stata datasets. as_df : bool Deprecated. Always returns a DataFrame Returns ------- dta : DataFrame A DataFrame containing the Stata dataset. Examples -------- >>> dta = webuse('auto') Notes ----- Make sure baseurl has trailing forward slash. Does not do any error checking in response URLs. z.dta)rr)databaseurlas_dfurls mC:\Users\julio\OneDrive\Documentos\Trabajo\Ideas Frescas\venv\Lib\site-packages\statsmodels/datasets/utils.pywebusers6 '; 'C c?c \rSrSrSrSrSrg)Dataset.c SUlSUlSUlSUl[R X5 XlURR[5Ul g! g=fN) endogexogrnamesdict__init____dict__astypefloatraw_data)selfkws rr$Dataset.__init__/sT     d   II,,U3DM  s $AA"c,[UR5$r)str __class__)r)s r__repr__Dataset.__repr__?s4>>""r)r%rr r!r"r(N)__name__ __module__ __qualname____firstlineno__r$r/__static_attributes__rrrr.s  #rrc URn[U[5(aAXAnXR5nUcUR U/SS9nOXUR5nOUR SS2U4R5n[ UR5nUcUR USS9nO@[U[5(aXUR5nOXUR5nUbJ[URSS2U45nXl UR5Ul URXC5n[ UR5n [U[ U5UXuU S9n U $)N)axis)rr"r r! endog_name exog_name) columns isinstanceintcopydroploclistrilocindex set_indexr) r endog_idxexog_idx index_idxr"r:r r!rDr;datasets rprocess_pandasrJCsE LLE)S!!%  %%'  99j\92Dh(--/DI&++-%--(  99Za90D # & &h(--/Dh(--/Ddii9 -. ZZ\ ~~e./T\\"I4tE{%)MG Nrc URR[[S[ U5S-555(aUR SS9nU$)zs All the Rdatasets have the integer row.labels from R if there is no real index. Strip this for a zero-based index r8T)r@)rDequalsrrlen reset_index)rs r_maybe_reset_indexrOcsE  zzvaTQ7899T* KrcRUSLaSnU$USLa [S5nU$[U5nU$)NFT) get_data_home)caches r _get_cacherSms@ ~ L $d# Le$ LrcSSKn[US5nURURU55 SSS5 g!,(df  g=f)Nrwb)zlibopenwritecompress)r cache_pathrVzfs r _cache_itr\xs2 j$ 2 t$%  s !; A cSSKn[US5nURUR55sSSS5 $!,(df  g=f)Nrrb)rVrW decompressread)rZrVr[s r _open_cachera~s/ j$ 2rwwy)  s : AcSnUbURS5SRSS5nURS5n[U5S:aUS==S - ss'O US ==S - ss'SRU5S -n[X5n[ U5nS nU(d'[ US S9R 5nUb [UW5 WU4$! N6=f)z Tries to load data from cache location otherwise downloads it. If it downloads the data and cache is not None then it will put the downloaded data in the cache path. Fz:///,.r8z-v2rz.zipT)timeout)splitreplacerMr rarr`r\)rrR from_cache file_namerZrs r_urlopen_cachedrns J IIe$R(00c: OOC( y>A  bMU "M aLE !LHHY'&0 %+  z*DJ sA&++-   dJ '    s  CCcXS-U--n[XB5upVUR SS5n[ U5U4$![a$nS[U5;a[SU-5eUeSnAff=f)Nz.%s404zDataset %s was not found.utf-8strict)rnr r- ValueErrordecoder)base_urldatanamerR extensionrrrlerrs r _get_datarys{ &)3 3C*36 ;;w )D D>: %%  CH 88CD DI s 8 A&A!!A&c dSn[X25upEURSS5n[[U55n[R "UR U:HURU:H5nUR5(d[SUSUSUS35eURUnUSRS $) NzRhttps://raw.githubusercontent.com/vincentarelbundock/Rdatasets/master/datasets.csvrqrrzItem z from Package z& was not found. Check the CSV file at z to verify the Item and Package.Titler) rnrtrrnp logical_andItemPackageanyrsrArC) rvpackagerR index_urlr_rDidx dataset_metas r_get_dataset_metars1Ii/GD ;;w )D Xd^ $E ..x/'1I JC 7799H:^G95(k)I K  99S>L  % %a ((rcSU-S-nSU-S-n[U5n[X0U5upV[USS9n[U5n[ XU5n[X@US5up[ XXR 5XUS9$) adownload and return R dataset Parameters ---------- dataname : str The name of the dataset you want to download package : str The package in which the dataset is found. The default is the core 'datasets' package. cache : bool or str If True, will download this data into the STATSMODELS_DATA folder. The default location is a folder called statsmodels_data in the user home folder. Otherwise, you can specify a path to a folder to use for caching the data. If False, the data will not be cached. Returns ------- dataset : Dataset A `statsmodels.data.utils.Dataset` instance. This objects has attributes: * data - A pandas DataFrame containing the data * title - The dataset title * package - The package from which the data came * from_cache - Whether not cached data was retrieved * __doc__ - The verbatim R documentation. Notes ----- If the R dataset has an integer index. This is reset to be zero-based. Otherwise the index is preserved. The caching facilities are dumb. That is, no download dates, e-tags, or otherwise identifying information is checked to see if the data should be downloaded again or not. If the dataset is in the cache, it's used. zJhttps://raw.githubusercontent.com/vincentarelbundock/Rdatasets/master/csv/rdzJhttps://raw.githubusercontent.com/vincentarelbundock/Rdatasets/master/doc/z/rst/r) index_colrst)r__doc__rtitlerl)rSryrrOrrr`) rvrrR data_base_url docs_base_urlrrlrdocrs r get_rdatasetrsJ##*++./M##*++23M u E %@D DA &D d #D h 7E }u =FC hhj'( **rcUc![R"S[SS55n[U5n[ U5(d [ U5 U$)aReturn the path of the statsmodels data dir. This folder is used by some large dataset loaders to avoid downloading the data several times. By default the data dir is set to a folder named 'statsmodels_data' in the user home folder. Alternatively, it can be set by the 'STATSMODELS_DATA' environment variable or programatically by giving an explicit folder path. The '~' symbol is expanded to the user home folder. If the folder does not already exist, it is automatically created. STATSMODELS_DATA~statsmodels_data)rgetr r r r data_homes rrQrQsIKK 2 $S*< =? 9%I )   rcF[U5n[R"U5 g)z.Delete all the content of the data home cache.N)rQshutilrmtreers rclear_data_homersi(I MM)rcXUcSOUn[U5 g![a nSnAgSnAff=f)zCheck if internet is availableNzhttps://github.comFT)rr )rrxs rcheck_internetrs4"%+ 3C   s  ))c/nUH}nURS5(aURS5(aUSSnO7URS5(aUSSnOURS5(aUSSnURU5 M XlU$)z Remove leading and trailing single quotes Parameters ---------- df : DataFrame DataFrame to process Returns ------- df : DataFrame DataFrame with stripped column names Notes ----- In-place modification 'r8rcN) startswithendswithappendr<)dfr<cs rstrip_column_namesr"s$G  <<  !**T"2"2!BA \\$  !"A ZZ  #2AqJ Irc[[U55n[XA5nUS:waSOSn0nUS:XaSS0n[U4X&S.UD6nU(aUR [ 5nU$)zStandard simple csv loaderrepythonrfloat_precisionhigh)sepengine)rrr rr&r') base_filecsv_namer convert_floatfilepathfilenamerrrs rload_csvrAshwy)*HH&HXFO },f5 H H# H HD{{5! Kr)z%https://www.stata-press.com/data/r11/T)rNN)csv)datasetsFr)reF),statsmodels.compat.pythonriorosrros.pathrrr r r r urllib.errorr r urllib.parserurllib.requestrnumpyr|pandasrrrrr#rrJrOrSr\rarnryrrrQrrrrr6rrrs, >> , "..>#d#*@& *< &)$2*n0 > r