9iH$SSKJrJrJrJrJr SSKrSSKr SSK J r SSK J r SSKJrJrJr SSKJrJrJr \"S5(aSSKr\"S 5(aSS KJr S r\R6"S S S \"5\"S5/S9r\R6"SS S S9r\R6"SSS \"\R<"S 5S9S9r\R6"SSS \"\R<"S5S9S9r \R6"SSSS9r!\R6"SSSS9r"\R6"SSSS9r#\R6"SSSS9r$\R6"SSS \ R "\RJ"S5S9\"S 5/S9r&\R6"S SS \ R "\RJ"S5S9\"S5/S9r'\RP\RR\RT\RV\RX\RZ\R\\R^\\\\ \!\"\#\$\&\'/r0S!\Rb4S"jr2S!\Rb4S#jr3S!\Rb4S$jr4S!\Rb4S%jr5S!\Rb4S&jr6S!\Rb4S'jr7S!\Rb4S(jr8S!\Rb4S)jr9S!\Rb4S*jr:S!\Rb4S+jr;S!\Rb4S,jr<\"S5S!\Rb4S-j5r=\"S5S!\Rb4S.j5r>S!\Rb4S/jr?S!\Rb4S0jr@\2\3\4\5\6\7\?\@\8\9\:\;\<\=\>S1.rASKS2\\ R\ R4S3\\DS4\\\\D\D4S5\\\D\E4S6\FS7\\\G\\D\G44S8\ R4S9jjrHSLS:\ RS;\\DS3\\DS<\\\\D\D4S=\\\\D\D4S>\\\\D\D4S?\\\\D\D4S@\\\\D\D4SA\\\\D\D4S5\\\D\E4SB\FSC\DSD\ES8\\D\\\ R\\D\I4444SEjjrJS8\\ R/\L44SFjrM\"S 5SMSG\S5\\\D\E4S?\\DSH\ESI\FS8\4 SJjj5rNg)N)CallableDictListOptionalUnionN)Version)holiday) DateOffsetDayEaster)function_requires_depsget_dep_versionis_dep_availablezchinese-calendarz scikit-learn)StandardScalerz Easter Sunday)monthdayoffsetz New Years Day)rr Superbowl)weekdayz Mothers DayzIndependence Day Christmas z New Years Evez Black Friday z Cyber MondayxcUR$N)yearr"s h/var/www/html/land-doc-ocr/venv/lib/python3.13/site-packages/paddlex/inference/models/common/ts/funcs.py _cal_yearr(M  66McUR$r$rr&s r' _cal_monthr-Ss  77Nr*cUR$r$rr&s r'_cal_dayr0Ys  55Lr*cUR$r$hourr&s r' _cal_hourr4_r)r*cUR$r$ dayofweekr&s r' _cal_weekdayr8es ;;r*cUR$r$)quarterr&s r' _cal_quarterr;ks 99r*c&URS- S- $)Ng7@?r2r&s r'_cal_hourofdayr>qs 66D=3 r*c&URS- S- $)Ng@r=r6r&s r'_cal_dayofweekr@ws ;; s ""r*c&URS- S- $)Ng>@r=r/r&s r'_cal_dayofmonthrB}s 554<# r*c&URS- S- $)Ngv@r=) dayofyearr&s r'_cal_dayofyearrEs ;;  $$r*c&URS- S- $)NgI@r=) weekofyearr&s r'_cal_weekofyearrHs <<$  $$r*c@[[R"U55$r$)floatchinese_calendar is_holidayr&s r' _cal_holidayrM !,,Q/ 00r*c@[[R"U55$r$)rJrK is_workdayr&s r' _cal_workdayrQrNr*c&URS- S- $)N;r=)minuter&s r'_cal_minuteofhourrUs 88b=3 r*c&URS- S- $)Ng&@r=r,r&s r'_cal_monthofyearrWs 77T>C r*)r%rrr3rr: minuteofhour monthofyear hourofdayr7 dayofmonthrDrGrLrPdatatime_col value_colsfreq drop_tail_nandtypereturnc\SnUc_[U[R5(aUR5nOPURSS2UR U:g4R5nO!URSS2U4R5nU(a=XR ;a[ SRU55eURSS2U4nO URn[R"UR[R5(a*[U[5(aUR[5n[R"UR[R5(a}U(a'[U[5(aUS:a [ S5eOSn[!U5[#U5U-pX- U- [%U5:wa [ S5e[R&"XUS9n O[R"UR[R(5(d4[R"UR[R*5(a[R,"USS9n[R."U5n U(a![U[5(d [ S 5eO>[R0"U 5nUc [ S 5eUS S :XaUSSnO [ S 5e[U[R5(aUR35nUR5U SS9 UR7SS9 U$)a?Transforms a DataFrame or Series into a time-indexed DataFrame. Args: data (Union[pd.DataFrame, pd.Series]): The input data containing time series information. time_col (Optional[str]): The column name representing time information. If None, uses the index. value_cols (Optional[Union[List[str], str]]): Columns to extract as values. If None, uses all except time_col. freq (Optional[Union[str, int]]): The frequency of the time series data. drop_tail_nan (bool): If True, drop trailing NaN values from the data. dtype (Optional[Union[type, Dict[str, type]]]): Enforce a specific data type on the resulting DataFrame. Returns: pd.DataFrame: A DataFrame with time as the index and specified value columns. Raises: ValueError: If the time column doesn't exist, or if frequency cannot be inferred. Nz0The time column: {} doesn't exist in the `data`!rzOThe type of `freq` should be `int` when the type of `time_col` is `RangeIndex`.z5The number of rows doesn't match with the RangeIndex!)startstopstepT)infer_datetime_formatzRThe type of `freq` should be `str` when the type of `time_col` is `DatetimeIndex`.z7Failed to infer the `freq`. A valid `freq` is required.r-z"The type of `time_col` is invalid.)inplace) isinstancepdSeriescopyloccolumns ValueErrorformatindexnp issubdtyperaintegerstrastypeintminmaxlen RangeIndexobject_ datetime64 to_datetime DatetimeIndex infer_freqto_frame set_index sort_index) r\r]r^r_r`ra series_data time_col_vals start_idxstop_idx time_indexs r'load_from_one_dataframers4K dBII & &))+K((1dllh&>#>?DDFKhhq*}-224  << 'BII(S H-   }}](("**55*T3:O:O%,,S1  }}](("**55 dC((D1H e-5 D!-0#m2Dt2K8  D (CI 5TU U]]M  }**BJJ 7 72==R]]<<}DQ %%m4 dC(( h) ==,D| MAw#~ABx=>>+ryy))!**, *d34( r*dfgroup_id target_cols label_colobserved_cov_cols feature_colsknown_cov_colsstatic_cov_colsfill_missing_dates fillna_methodfillna_window_sizec /nUbBXR5nUH)nURXURU/55 M+ OU/n/nU(a1[U[5(a[ U5S:a [ S5eUnU(aUnUGH+nSnSnSn[5n[X5Xx/5(d2[UUURVs/sH nUU:wdM UPM snU 5nOU(a[UUUU 5nU(a[UUUU 5nU(a[UUUU 5nU(aw[U[5(aU/nUHYnUUR;d&[ [R"UU55S:wa [ S5eUURSUU'M[ URUUUUS.5 GM. US$s snf)aiLoads and processes time series data from a DataFrame. This function extracts and organizes time series data from a given DataFrame. It supports optional grouping and extraction of specific columns as features. Args: df (pd.DataFrame): The input DataFrame containing time series data. group_id (Optional[str]): Column name used for grouping the data. time_col (Optional[str]): Name of the time column. target_cols (Optional[Union[List[str], str]]): Columns to be used as target. label_col (Optional[Union[List[str], str]]): Columns to be used as label. observed_cov_cols (Optional[Union[List[str], str]]): Columns for observed covariates. feature_cols (Optional[Union[List[str], str]]): Columns to be used as features. known_cov_cols (Optional[Union[List[str], str]]): Columns for known covariates. static_cov_cols (Optional[Union[List[str], str]]): Columns for static covariates. freq (Optional[Union[str, int]]): Frequency of the time series data. fill_missing_dates (bool): Whether to fill missing dates in the time series. fillna_method (str): Method to fill missing values ('pre' or 'post'). fillna_window_size (int): Window size for filling missing values. **kwargs: Additional keyword arguments. Returns: Dict[str, Optional[Union[pd.DataFrame, Dict[str, any]]]]: A dictionary containing processed time series data. Nrz"The length of label_col must be 1.zIStatic covariate columns data is not in columns or schema is not correct!r) past_targetobserved_cov_numericknown_cov_numericstatic_cov_numeric) uniqueappendisinrjrvr{rpdictanyrrorsiloc)rrr]rrrrrrr_rrrkwargsdfs group_uniquecolumnrestarget observed_cov known_cov static_covacols r'load_from_dataframersR C|**, "F JJrX,++VH56 7#d C i % %#i.1*<AB B (  V KNTUU,JJ8Jq!x-J8 F0 !6%   3"  os33'6&7O*C"**,BIIbg4F0G10L(g')gll1oJsO + %(4%.&0   ct q6M_9s GGcD^S[RS[4U4SjjnU$)aCreates a function to calculate the distance in days to the nearest holiday. This function generates a closure that computes the number of days from a given date index to the nearest holiday within a defined window. Args: holiday: An object that provides a `dates` method, which returns the dates of holidays within a specified range. Returns: Callable[[pd.Timestamp], float]: A function that takes a date index as input and returns the distance in days to the nearest holiday. rrrbc>TRU[R"[S9- U[R"[S9-5n[ U5S:wd SUS35e[ XS- R 5$)a]Calculates the distance in days from a given date index to the nearest holiday. Args: index (pd.Timestamp): The date index for which the distance to the nearest holiday should be calculated. Returns: float: The number of days to the nearest holiday. Raises: AssertionError: If no holiday is found within the specified window. )daysrz&No closest holiday for the date index z found.)datesrk Timedelta MAX_WINDOWr{rJr)rr holiday_dater s r'_distance_to_day._distance_to_holiday.._distance_to_daysy}} BLLj1 1 BLLj1 1   " C 3E7' B C "e1o-3344r*)rk TimestamprJ)r rs` r'_distance_to_holidayrs"5 550 r*dataset extend_pointsric d^UnU(dUR5nUSnU(dUSRR5nOURR5nURSn[R "XxR [R5(a [S5eU(dUbUO[R"Xx5n[S5n [U 5[S5:a/[R"XxSUUS-S US 9R5n O.[R"XxSUUS-S US 9R5n [R"Xz/5nUGHmTS :waXxRU4S j5n XxU lUSc3[R "U R#T5U RS9US'McU R#T5R%USR5UST'M/n ['[(5HupXxR[+U55n XxU lU R-TS-[/U 5-5 UScB[R "U R#TS-[/U 5-5U RS9US'MU R#T5R%USR5USTS-[/U 5-'M [15nUR3USU 5 UR5USU 5USU 'GM U$)a#Transforms the time column of a dataset into time features. This function extracts time-related features from the time column in a dataset, optionally extending the time series for future points and normalizing holiday distances. Args: dataset (Dict): Dataset to be transformed. freq: Optional[Union[str, int]]: Frequency of the time series data. If not provided, the frequency will be inferred. feature_cols (List[str]): List of feature columns to be extracted. extend_points (int): Number of future points to extend the time series. inplace (bool): Whether to perform the transformation inplace. Default is False. Returns: Dict: The transformed dataset with time features added. Raises: ValueError: If the time column is of an integer type instead of datetime. rrrz\The time_col can't be the type of numpy.integer, and it must be the type of numpy.datetime64pandasz1.4rright)rdr_periods inclusivename)rdr_rclosedrholidaysc">[T"U5$r$)CAL_DATE_METHOD)r"ks r'time_feature.. s/!2DQ2Gr*)rr_)rmrrrrorsrtrarurprkrrr date_rangeconcatapply DataFramerenamereindex enumerateHOLIDAYSrrrvrfit transform)rr_rrrinew_tskcovtf_kcovr] pd_version extend_timev holidays_coliHscalerrs @r' time_featurers8F  % &D '--668**%%'q!H }}W&,,bjj99 j   'tR]]7;L-M$X. : '%. 0--'+%)!  hj --'+%)  hj ))W23 ?!''(GHA'AG)*2.0ll188A;agg.V*+12!1D1D./552*+A.L!(+%++,@,CD!+##AGc!f$45-.624,,S3q6!12!''3F./EFHHQKDWDW2399EF./C#a&0@A,$%F JJv12<@ A8>8H8H*+L99F& ' 59> Mr*)NNNFN) NNNNNNNNNFpre )F)Otypingrrrrrnumpyrsrrkpackaging.versionrpandas.tseriesr hdpandas.tseries.offsetsr r r utils.depsrrrrKsklearn.preprocessingrrHoliday EasterSunday NewYearsDaySU SuperBowl MothersDayIndependenceDay ChristmasEve ChristmasDay NewYearsEveTH BlackFriday CyberMonday EasterMonday GoodFriday USColumbusDay USLaborDayUSMartinLutherKingJr USMemorialDayUSPresidentsDayUSThanksgivingDayrr~r(r-r0r4r8r;r>r@rBrErHrMrQrUrWrrrlrvrxbooltyperrrrrJrrr*r'rs 98%(::UU&''N##4  zz/68SQRVBTU jjq9 JJ{!:beeTUh;W X ZZ*RUU1X*F **/qa@zz+RR8 zz+RR8 jj; jj  MM"%%( +SV 4  jj  MM"%%( +SV 4  OOMMMM % , }}  }}  }}  }}  }}  }}  }} # }}#  }} % }}% % }}% *+1 }}1,1 *+1 }}1,1  }}   }}     %#!!*#26&*48 _  bii' (_sm_tCy#~./_ 5c? # _  _ E$S$Y/0 1 _\\_H#"37159=486:7;&*$ { {sm{sm{%S 3/0 { d3in-. {  d3in 56 {5cC01{U49c>23{eDIsN34{ 5c? #{{{{ #xbllDcN:;< <={|'Xr||ne.C%D'T'  ` ` 5c? #`s)` `  `  `(`r*