upgini 1.1.280.dev0__py3-none-any.whl → 1.2.31a2__py3-none-any.whl

upgini/normalizer/phone_normalizer.py

@@ -1,340 +0,0 @@
-from typing import Optional
-
-import pandas as pd
-from pandas.api.types import is_float_dtype, is_int64_dtype, is_object_dtype, is_string_dtype
-
-from upgini.errors import ValidationError
-
-
-class PhoneNormalizer:
-    def __init__(self, df: pd.DataFrame, phone_column_name: str, country_column_name: Optional[str] = None):
-        self.df = df
-        self.phone_column_name = phone_column_name
-        self.country_column_name = country_column_name
-
-    def normalize(self) -> pd.DataFrame:
-        self.phone_to_int()
-        if self.country_column_name is not None:
-            self.df = self.df.apply(self.add_prefix, axis=1)
-        return self.df[self.phone_column_name].astype("Int64")
-
-    def add_prefix(self, row):
-        phone = row[self.phone_column_name]
-        if pd.isna(phone):
-            return row
-        country = row[self.country_column_name]
-        country_prefix_tuple = self.COUNTRIES_PREFIXES.get(country)
-        if country_prefix_tuple is not None:
-            country_prefix, number_of_digits = country_prefix_tuple
-            if len(str(phone)) == number_of_digits:
-                row[self.phone_column_name] = int(country_prefix + str(phone))
-        return row
-
-    def phone_to_int(self):
-        """
-        Convention: phone number is always presented as int number.
-        phone_number = Country code + National Destination Code + Subscriber Number.
-        Examples:
-        41793834315     for Switzerland
-        46767040672     for Sweden
-        861065529988    for China
-        18143008198     for the USA
-        Inplace conversion of phone to int.
-
-        Method will remove all non numeric chars from string and convert it to int.
-        None will be set for phone numbers that couldn"t be converted to int
-        """
-        if is_string_dtype(self.df[self.phone_column_name]) or is_object_dtype(self.df[self.phone_column_name]):
-            convert_func = self.phone_str_to_int_safe
-        elif is_float_dtype(self.df[self.phone_column_name]):
-            convert_func = self.phone_float_to_int_safe
-        elif is_int64_dtype(self.df[self.phone_column_name]):
-            convert_func = self.phone_int_to_int_safe
-        else:
-            raise ValidationError(
-                f"phone_column_name {self.phone_column_name} doesn't have supported dtype. "
-                f"Dataset dtypes: {self.df.dtypes}. "
-                f"Contact developer and request to implement conversion of {self.phone_column_name} to int"
-            )
-        self.df[self.phone_column_name] = self.df[self.phone_column_name].apply(convert_func).astype("Int64")
-
-    @staticmethod
-    def phone_float_to_int_safe(value: float) -> Optional[int]:
-        try:
-            return PhoneNormalizer.validate_length(int(value))
-        except Exception:
-            return None
-
-    @staticmethod
-    def phone_int_to_int_safe(value: int) -> Optional[int]:
-        try:
-            return PhoneNormalizer.validate_length(int(value))
-        except Exception:
-            return None
-
-    @staticmethod
-    def phone_str_to_int_safe(value: str) -> Optional[int]:
-        try:
-            value = str(value)
-            if value.endswith(".0"):
-                value = value[: len(value) - 2]
-            numeric_filter = filter(str.isdigit, value)
-            numeric_string = "".join(numeric_filter)
-            return PhoneNormalizer.validate_length(int(numeric_string))
-        except Exception:
-            return None
-
-    @staticmethod
-    def validate_length(value: int) -> Optional[int]:
-        if value < 10000000 or value > 999999999999999:
-            return None
-        else:
-            return value
-
-    COUNTRIES_PREFIXES = {
-        "US": ("1", 10),
-        "CA": ("1", 10),
-        "AI": ("1", 10),
-        "AG": ("1", 10),
-        "AS": ("1", 10),
-        "BB": ("1", 10),
-        "BS": ("1", 10),
-        "VG": ("1", 10),
-        "VI": ("1", 10),
-        "KY": ("1", 10),
-        "BM": ("1", 10),
-        "GD": ("1", 10),
-        "TC": ("1", 10),
-        "MS": ("1", 10),
-        "MP": ("1", 10),
-        "GU": ("1", 10),
-        "SX": ("1", 10),
-        "LC": ("1", 10),
-        "DM": ("1", 10),
-        "VC": ("1", 10),
-        "PR": ("1", 10),
-        "TT": ("1", 10),
-        "KN": ("1", 10),
-        "JM": ("1", 10),
-        "EG": ("20", 9),
-        "SS": ("211", 9),
-        "MA": ("212", 9),
-        "EH": ("212", 4),
-        "DZ": ("213", 8),
-        "TN": ("216", 8),
-        "LY": ("218", 9),
-        "GM": ("220", 6),
-        "SN": ("221", 9),
-        "MR": ("222", 7),
-        "ML": ("223", 8),
-        "GN": ("224", 9),
-        "CI": ("225", 7),
-        "BF": ("226", 8),
-        "NE": ("227", 8),
-        "TG": ("228", 8),
-        "BJ": ("229", 8),
-        "MU": ("230", 7),
-        "LR": ("231", 9),
-        "SL": ("232", 8),
-        "GH": ("233", 9),
-        "NG": ("234", 9),
-        "TD": ("235", 8),
-        "CF": ("236", 7),
-        "CM": ("237", 9),
-        "CV": ("238", 7),
-        "ST": ("239", 7),
-        "GQ": ("240", 9),
-        "GA": ("241", 8),
-        "CG": ("242", 7),
-        "CD": ("243", 9),
-        "AO": ("244", 9),
-        "GW": ("245", 6),
-        "IO": ("246", 7),
-        "AC": ("247", 5),
-        "SC": ("248", 7),
-        "SD": ("249", 9),
-        "RW": ("250", 9),
-        "ET": ("251", 9),
-        "SO": ("252", 9),
-        "DJ": ("253", 8),
-        "KE": ("254", 9),
-        "TZ": ("255", 9),
-        "UG": ("256", 9),
-        "BI": ("257", 8),
-        "MZ": ("258", 8),
-        "ZM": ("260", 9),
-        "MG": ("261", 9),
-        "RE": ("262", 9),
-        "YT": ("262", 9),
-        "TF": ("262", 9),
-        "ZW": ("263", 9),
-        "NA": ("264", 9),
-        "MW": ("265", 7),
-        "LS": ("266", 8),
-        "BW": ("267", 7),
-        "SZ": ("268", 8),
-        "KM": ("269", 7),
-        "ZA": ("27", 10),
-        "SH": ("290", 5),
-        "TA": ("290", 5),
-        "ER": ("291", 7),
-        "AT": ("43", 10),
-        "AW": ("297", 7),
-        "FO": ("298", 6),
-        "GL": ("299", 6),
-        "GR": ("30", 10),
-        "BE": ("32", 8),
-        "FR": ("33", 9),
-        "ES": ("34", 9),
-        "GI": ("350", 8),
-        "PE": ("51", 8),
-        "MX": ("52", 10),
-        "CU": ("53", 8),
-        "AR": ("54", 10),
-        "BR": ("55", 10),
-        "CL": ("56", 9),
-        "CO": ("57", 8),
-        "VE": ("58", 10),
-        "PT": ("351", 9),
-        "LU": ("352", 8),
-        "IE": ("353", 8),
-        "IS": ("354", 7),
-        "AL": ("355", 8),
-        "MT": ("356", 8),
-        "CY": ("357", 8),
-        "FI": ("358", 9),
-        "BG": ("359", 8),
-        "HU": ("36", 8),
-        "LT": ("370", 8),
-        "LV": ("371", 8),
-        "EE": ("372", 7),
-        "MD": ("373", 8),
-        "AM": ("374", 8),
-        "BY": ("375", 9),
-        "AD": ("376", 6),
-        "MC": ("377", 8),
-        "SM": ("378", 9),
-        "VA": ("3906698", 5),
-        "UA": ("380", 9),
-        "RS": ("381", 9),
-        "ME": ("382", 8),
-        "HR": ("385", 8),
-        "SI": ("386", 8),
-        "BA": ("387", 8),
-        "MK": ("389", 8),
-        "MY": ("60", 9),
-        "AU": ("61", 9),
-        "CX": ("61", 9),
-        "CC": ("61", 9),
-        "ID": ("62", 9),
-        "PH": ("632", 7),
-        "NZ": ("64", 8),
-        "PN": ("64", 8),
-        "SG": ("65", 8),
-        "TH": ("66", 8),
-        "IT": ("39", 10),
-        "RO": ("40", 9),
-        "CH": ("41", 9),
-        "CZ": ("420", 9),
-        "SK": ("421", 9),
-        "GB": ("44", 10),
-        "LI": ("423", 7),
-        "GG": ("44", 10),
-        "IM": ("44", 10),
-        "JE": ("44", 10),
-        "DK": ("45", 8),
-        "SE": ("46", 8),
-        "BD": ("880", 8),
-        "TW": ("886", 9),
-        "JP": ("81", 9),
-        "KR": ("82", 9),
-        "VN": ("84", 10),
-        "KP": ("850", 8),
-        "HK": ("852", 8),
-        "MO": ("853", 8),
-        "KH": ("855", 8),
-        "LA": ("856", 8),
-        "NO": ("47", 8),
-        "SJ": ("47", 8),
-        "BV": ("47", 8),
-        "PL": ("48", 9),
-        "DE": ("49", 10),
-        "TR": ("90", 10),
-        "IN": ("91", 10),
-        "PK": ("92", 9),
-        "AF": ("93", 9),
-        "LK": ("94", 9),
-        "MM": ("95", 7),
-        "IR": ("98", 10),
-        "MV": ("960", 7),
-        "LB": ("961", 7),
-        "JO": ("962", 9),
-        "SY": ("963", 10),
-        "IQ": ("964", 10),
-        "KW": ("965", 7),
-        "SA": ("966", 9),
-        "YE": ("967", 7),
-        "OM": ("968", 8),
-        "PS": ("970", 8),
-        "AE": ("971", 8),
-        "IL": ("972", 9),
-        "BH": ("973", 8),
-        "QA": ("974", 8),
-        "BT": ("975", 7),
-        "MN": ("976", 8),
-        "NP": ("977", 8),
-        "TJ": ("992", 9),
-        "TM": ("993", 8),
-        "AZ": ("994", 9),
-        "GE": ("995", 9),
-        "KG": ("996", 9),
-        "UZ": ("998", 9),
-        "FK": ("500", 5),
-        "BZ": ("501", 7),
-        "GT": ("502", 8),
-        "SV": ("503", 8),
-        "HN": ("504", 8),
-        "NI": ("505", 8),
-        "CR": ("506", 8),
-        "PA": ("507", 7),
-        "PM": ("508", 6),
-        "HT": ("509", 8),
-        "GS": ("500", 5),
-        "MF": ("590", 9),
-        "BL": ("590", 9),
-        "GP": ("590", 9),
-        "BO": ("591", 9),
-        "GY": ("592", 9),
-        "EC": ("593", 9),
-        "GF": ("594", 9),
-        "PY": ("595", 9),
-        "MQ": ("596", 9),
-        "SR": ("597", 9),
-        "UY": ("598", 9),
-        "CW": ("599", 9),
-        "BQ": ("599", 9),
-        "RU": ("7", 10),
-        "KZ": ("7", 10),
-        "TL": ("670", 7),
-        "NF": ("672", 7),
-        "HM": ("672", 7),
-        "BN": ("673", 7),
-        "NR": ("674", 7),
-        "PG": ("675", 7),
-        "TO": ("676", 7),
-        "SB": ("677", 7),
-        "VU": ("678", 7),
-        "FJ": ("679", 7),
-        "PW": ("680", 7),
-        "WF": ("681", 7),
-        "CK": ("682", 5),
-        "NU": ("683", 7),
-        "WS": ("685", 7),
-        "KI": ("686", 7),
-        "NC": ("687", 7),
-        "TV": ("688", 7),
-        "PF": ("689", 7),
-        "TK": ("690", 7),
-        "FM": ("691", 7),
-        "MH": ("692", 7),
-    }

upgini-1.1.280.dev0.dist-info/RECORD

@@ -1,62 +0,0 @@
-upgini/__about__.py,sha256=UrHon9iXf0_L0ezeavtXrVCL8ABnaX0u5iXp6FC3PTA,29
-upgini/__init__.py,sha256=asENHgEVHQBIkV-e_0IhE_ZWqkCG6398U3ZLrNzAH6k,407
-upgini/ads.py,sha256=nvuRxRx5MHDMgPr9SiU-fsqRdFaBv8p4_v1oqiysKpc,2714
-upgini/dataset.py,sha256=uiFY-P8te7-zigib1hGWRtW5v0X7chxPM0hJFdixAN8,45623
-upgini/errors.py,sha256=2b_Wbo0OYhLUbrZqdLIx5jBnAsiD1Mcenh-VjR4HCTw,950
-upgini/features_enricher.py,sha256=HQHTKpyK9MGqYTZ5FMgJLvRQ0o1JwApcQklTRKxIgMs,176683
-upgini/http.py,sha256=khrYSldpY-HbVLCcApfV1BjBFK6Uyuatb4colKybxgY,42301
-upgini/metadata.py,sha256=CFJekYGD7Ep7pRFH7wCEcsXS4bz83do33FNmtcCY9P4,9729
-upgini/metrics.py,sha256=L4LKSMOK9iKFLaJvTBTKk2tQauMgiJqtfrBclM3fBjs,29670
-upgini/search_task.py,sha256=LtRJ9bCPjMo1gJ-sUDKERhDwGcWKImrzwVFHjkMSQHQ,17071
-upgini/spinner.py,sha256=4iMd-eIe_BnkqFEMIliULTbj6rNI2HkN_VJ4qYe0cUc,1118
-upgini/version_validator.py,sha256=RGg87VweujTNlibgsOuqPLIEiBgIOkuXNVTGuNCD234,1405
-upgini/ads_management/__init__.py,sha256=qzyisOToVRP-tquAJD1PblZhNtMrOB8FiyF9JvfkvgE,50
-upgini/ads_management/ads_manager.py,sha256=igVbN2jz80Umb2BUJixmJVj-zx8unoKpecVo-R-nGdw,2648
-upgini/autofe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-upgini/autofe/all_operands.py,sha256=SyKVU-xGMHgoRZvHrCmba2u2Ygc73c1mXFolNSWe8Uo,2357
-upgini/autofe/binary.py,sha256=441BRuqMsxlxuw4c8rMZB6h5EpRdVMk-bVa03U7T5Hg,3973
-upgini/autofe/date.py,sha256=Vy1I92fLLYLhuYKJmtuPBMI8cPxE4Uwk40hqE2F2e1A,4224
-upgini/autofe/feature.py,sha256=ChSuuIbRPGIWnPjKAgZbeAEi7Y_PjSVRyxxx41MyFp0,11845
-upgini/autofe/groupby.py,sha256=4WjDzQxqpZxB79Ih4ihMMI5GDxaFqiH6ZelfV82ClT4,3091
-upgini/autofe/operand.py,sha256=xgEIZuFCfckc6LpBqVu1OVK3JEabm1O-LHUsp83EHKA,2806
-upgini/autofe/unary.py,sha256=v-l3aiE5hj6kurvh6adCQL8W3X9u9a7RVbS_WPR2qlw,3146
-upgini/autofe/vector.py,sha256=dLxfAstJs-gw_OQ1xxoxcM6pVzORlV0HVzdzt7cLXVQ,606
-upgini/data_source/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-upgini/data_source/data_source_publisher.py,sha256=taRzyGgrPrTTSGw4Y-Ca5k4bf30aiTa68rxqT9zfqeI,16478
-upgini/mdc/__init__.py,sha256=aM08nIWFc2gWdWUa3_IuEnNND0cQPkBGnYpRMnfFN8k,1019
-upgini/mdc/context.py,sha256=3u1B-jXt7tXEvNcV3qmR9SDCseudnY7KYsLclBdwVLk,1405
-upgini/normalizer/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-upgini/normalizer/phone_normalizer.py,sha256=EzTaahk6myRv6ZXgbyVFGY4kpo_2VlQgOrm5_lfbmNI,9996
-upgini/resource_bundle/__init__.py,sha256=S5F2G47pnJd2LDpmFsjDqEwiKkP8Hm-hcseDbMka6Ko,8345
-upgini/resource_bundle/exceptions.py,sha256=5fRvx0_vWdE1-7HcSgF0tckB4A9AKyf5RiinZkInTsI,621
-upgini/resource_bundle/strings.properties,sha256=1O779a0-Ai0j7W-Z5AznvjuV69YkJvgGhJda-6VMLOQ,26287
-upgini/resource_bundle/strings_widget.properties,sha256=gOdqvZWntP2LCza_tyVk1_yRYcG4c04K9sQOAVhF_gw,1577
-upgini/sampler/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-upgini/sampler/base.py,sha256=7GpjYqjOp58vYcJLiX__1R5wjUlyQbxvHJ2klFnup_M,6389
-upgini/sampler/random_under_sampler.py,sha256=TIbm7ATo-bCMF-IiS5sZeDC1ad1SYg0eY_rRmg84yIQ,4024
-upgini/sampler/utils.py,sha256=PYOk3kKSnFlyxcpdtDNLBEEhTB4lO_iP7pQHqeUcmAc,20211
-upgini/utils/__init__.py,sha256=O_KgzKiJjW3g4NoqZ7lAxUpoHcBi_gze6r3ndEjCH74,842
-upgini/utils/base_search_key_detector.py,sha256=UNs2uxEcD1N_mOtkx3k6U70DCajW-QEO2vZp41GF0mU,855
-upgini/utils/blocked_time_series.py,sha256=Uqr3vp4YqNclj2-PzEYqVy763GSXHn86sbpIl1UOB4s,3382
-upgini/utils/country_utils.py,sha256=yE8oRgMpXuJxPfQm4fioY6dg6700HgVnHSk4Cv9sUyM,6511
-upgini/utils/custom_loss_utils.py,sha256=bLk3uygqkJBaGkyzfO032d72QASae-dDyEURfFehVJo,3973
-upgini/utils/cv_utils.py,sha256=w6FQb9nO8BWDx88EF83NpjPLarK4eR4ia0Wg0kLBJC4,3525
-upgini/utils/datetime_utils.py,sha256=-LsDTThsGKsTZ57V1uNiHtLcoTtqktk5tui4WnqggJo,10673
-upgini/utils/deduplicate_utils.py,sha256=6AbARehUCghJZ4PppFtrej2s3gFRruh41MEm6mzakHs,8607
-upgini/utils/display_utils.py,sha256=LKoSwjrE0xgS5_cqVhc2og2CQ1UCZ1nTI2VKboIhoQA,10858
-upgini/utils/email_utils.py,sha256=PLufTO97Pg9PPsNqB9agcM6M98MIxKUgIgNn2mVwSQ0,3520
-upgini/utils/fallback_progress_bar.py,sha256=PDaKb8dYpVZaWMroNcOHsTc3pSjgi9mOm0--cOFTwJ0,1074
-upgini/utils/features_validator.py,sha256=PgKNt5dyqfErTvjtRNNUS9g7GFqHBtAtnsfA-V5UO1A,3307
-upgini/utils/format.py,sha256=Yv5cvvSs2bOLUzzNu96Pu33VMDNbabio92QepUj41jU,243
-upgini/utils/ip_utils.py,sha256=Zf3F2cnQmOCH09QLQHetpjMFu1PnD0cTmDymn0SnSy8,1672
-upgini/utils/phone_utils.py,sha256=JNSkF8G6mgsN8Czy11pamaJdsY6rBINEMpi7jbVt_RA,408
-upgini/utils/postal_code_utils.py,sha256=_8CR9tBqsPptQsmMUvnrCAmBaMIQSWH3JfJ4ly3x_zs,409
-upgini/utils/progress_bar.py,sha256=N-Sfdah2Hg8lXP_fV9EfUTXz_PyRt4lo9fAHoUDOoLc,1550
-upgini/utils/sklearn_ext.py,sha256=c23MGSUVfxLnaDWKAxavHgnOtm5dGKkF3YswdWQcFzs,43984
-upgini/utils/target_utils.py,sha256=Y96_PJ5cC-WsEbeqg20v9uqywDQobLoTb-xoP7S3o4E,7807
-upgini/utils/track_info.py,sha256=G5Lu1xxakg2_TQjKZk4b5SvrHsATTXNVV3NbvWtT8k8,5663
-upgini/utils/warning_counter.py,sha256=dIWBB4dI5XRRJZudvIlqlIYKEiwLLPcXarsZuYRt338,227
-upgini-1.1.280.dev0.dist-info/METADATA,sha256=AlD_LoKBQRcf5DSTcQeTung2oC2xtzCzKnBqgMeVUH0,48123
-upgini-1.1.280.dev0.dist-info/WHEEL,sha256=zEMcRr9Kr03x1ozGwg5v9NQBKn3kndp6LSoSlVg-jhU,87
-upgini-1.1.280.dev0.dist-info/licenses/LICENSE,sha256=5RRzgvdJUu3BUDfv4bzVU6FqKgwHlIay63pPCSmSgzw,1514
-upgini-1.1.280.dev0.dist-info/RECORD,,