File size: 37,967 Bytes
8643b59 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 | """
@author: Edward R Jones
@version 1.34
@copyright 2020 - Edward R Jones, all rights reserved.
"""
#from DT import DT
import sys
import warnings
import numpy as np
import pandas as pd
from sklearn import preprocessing
from sklearn.impute import SimpleImputer
from copy import deepcopy #Used to create sentiment word dictionary
import pickle
from enum import Enum
#Class DT - DataType This is setup to provide a clean
#notation for data maps used by ReplaceImputeEncode
class DT(Enum):
# @attributes: characters recognized in RIE code
Interval = 'I' #Expected values (lowest value, highest value)
Binary = 'B' #Expected values (class0, class1)
Nominal = 'N' #Expected values (class0, class1, ... classk)
Ordinal = 'O' #Expected values ordered classes (class0, class1, ...)
String = 'S' #Expected values ("")
ID = 'Z' #Expected values ("")
Label = 'L' #Expected values ("")
Text = 'T' #Expected values ("")
Ignore = 'Z' #Expected values ("")
interval = 'I' #Allow lower case
binary = 'B' #Allow lower case
nominal = 'N' #Allow lower case
ordinal = 'O' #Allow lower case
string = 'S' #Allow lower case
id = 'Z' #Expected values ("")
label = 'L' #Expected values ("")
text = 'T' #Allow lower case
ignore = 'Z' #Allow lower case
# @methods
def getDataTypes():
dtype = [
DT.Interval,
DT.Binary,
DT.Nominal,
DT.Ordinal,
DT.ID,
DT.Label,
DT.Text ,
DT.String,
DT.Ignore
]
return dtype #Returns data type list
def convertDataType(atype):
if atype==DT.Interval:
ctype ='DT.Interval'
elif atype==DT.Binary:
ctype ='DT.Binary'
elif atype==DT.Nominal:
ctype ='DT.Nominal'
elif atype==DT.Ordinal:
ctype ='DT.Ordinal'
elif atype==DT.String:
ctype ='DT.String'
elif atype==DT.ID:
ctype ='DT.ID'
elif atype==DT.Label:
ctype ='DT.Label'
elif atype==DT.Text:
ctype ='DT.Text'
else:ctype ='DT.Ignore'
return ctype
"""
class ReplaceImputeEncode
@parameters:
*** __init__() ***
data_map - The metadata dictionary.
nominal_encoding - Can be 'one-hot', 'SAS' or default None.
interval_scale - Can be 'std', 'robust' or default None.
no_impute - default None or list of attributes to exclude from
imputation
drop - True or default False. True drops the last nominal
encoded column. False keeps all nominal encoded
columns.
display - True or default False. True displays the number of
missing and outliers found in the data.
*** fit_transform () ***
df - a pandas DataFrame containing the data description
by the metadata found in data_map (required)
data_map - See above description.
@Cautions:
The incoming dataframe, df, and the data_map are deep copied to
ensure that changes to the dataframe are only held within the class
object self.copy_df. The attributes_map is deep copied into
self.features_map. All binary and nominal values are encoded to
numeric values.
The method draft_data_map returns a of the data_map based upon
the data. This must be examined to ensure the data types and
allowed values are correct. This behavior is controled by k_min
and k_max. See API and examples for details.
"""
class ReplaceImputeEncode(object):
def __init__(self, data_map=None, binary_encoding=None,
nominal_encoding=None, interval_scale=None, no_impute=None,
no_encode=None, drop=False, display=False):
if interval_scale=='None' or interval_scale=='none':
self.interval_scale=None
else:
self.interval_scale=interval_scale
self.go_flag = False
self.features_map = data_map
self.drop = drop
self.display = display
self.interval_scale = interval_scale
self.no_impute = no_impute
self.no_encode = no_encode
if binary_encoding=='None' or binary_encoding=='none':
self.binary_encoding = None
else:
self.binary_encoding = binary_encoding
#nominal_encoding can be 'SAS' or 'one-hot'
if binary_encoding != 'SAS' and binary_encoding != 'one-hot' \
and binary_encoding != None:
raise ValueError("***Call to ReplaceImputeEncode invalid. "+
"*** binary_encoding="+binary_encoding+" is invalid."+
"*** must use None, 'one-hot' or 'SAS'")
sys.exit()
if nominal_encoding=='None' or nominal_encoding=='none':
self.nominal_encoding = None
else:
self.nominal_encoding = nominal_encoding
#nominal_encoding can be 'SAS' or 'one-hot'
if nominal_encoding != 'SAS' and nominal_encoding != 'one-hot' \
and nominal_encoding != None:
raise ValueError("***Call to ReplaceImputeEncode invalid. "+
"*** nominal_encoding="+nominal_encoding+" is invalid."+
"*** must use None, 'one-hot' or 'SAS'")
sys.exit()
if interval_scale != 'std' and interval_scale != 'robust' \
and interval_scale != None:
raise ValueError("***Call to ReplaceImputeEncode invalid. "+
"*** interval_scale="+interval_scale+" is invalid."+
"*** must use None, 'std' or 'robust'")
sys.exit()
if data_map==None:
print("Attributes Map is required.")
print("Please pass map using data_map attribute.")
print("If one is not available, try creating one using "+
"call to draft_data_map(df)")
return
if type(data_map)==str:
try:
self.features_map = self.load_data_map(data_map)
except:
raise ValueError("Unable to load data map:", data_map)
sys.exit()
elif type(data_map)==dict:
self.features_map = data_map
else:
raise ValueError("Supplied Data Map not Dictionary or File")
sys.exit()
self.interval_attributes = []
self.nominal_attributes = []
self.binary_attributes = []
self.onehot_attributes = []
self.onehot_cats = []
self.hot_drop_list = []
self.missing_counts = {}
self.outlier_counts = {}
for feature,v in self.features_map.items():
# Initialize data map missing and outlier counters to zero
self.missing_counts[feature] = 0
self.outlier_counts[feature] = 0
if v[0] not in DT.getDataTypes():
raise TypeError(
"\n***Data Map in call to ReplaceImputeEncode invalid.\n"+
"***Data Type for '"+ feature + "' is not recognized. "+
"\n***Valid types are: DT.Interval, DT.Binary, DT.Nominal, "+
"DT.Text, DT.String, DT.ID, or DT.Ignore")
if v[0]==DT.Interval:
self.interval_attributes.append(feature)
else:
if v[0]==DT.Binary:
self.binary_attributes.append(feature)
else:
if v[0]!=DT.Binary and v[0]!=DT.Nominal:
# Ignore, don't touch this attribute
continue
# Attribute must be Nominal
self.nominal_attributes.append(feature)
# Setup column names for Nominal encoding
n_cat = len(v[1])
self.onehot_cats.append(list(v[1]))
data_type = type(v[1][n_cat-1])
if self.drop == True:
n_cat -= 1
for i in range(n_cat):
if type(v[1][i]) != data_type:
raise TypeError(
"\n***Classes invalid for--> '"+feature+"'"+
"\n***Must be all numeric or strings, not both.")
if type(v[1][i])==int:
my_str = feature+str(v[1][i])
else:
my_str = feature+("%i" %i)+":"+str(v[1][i])[0:10]
self.onehot_attributes.append(my_str)
self.n_interval = len(self.interval_attributes)
self.n_binary = len(self.binary_attributes)
self.n_nominal = len(self.nominal_attributes)
self.n_onehot = len(self.onehot_attributes)
self.cat = self.n_binary + self.n_nominal
if nominal_encoding=='SAS' and drop==False and self.n_nominal>0:
raise ValueError("***Call to ReplaceImputeEncode invalid. "+
"***nominal_encoding='SAS' requested with drop=False "+
"***'SAS' encoding requires drop=True")
sys.exit()
self.col = []
for i in range(self.n_interval):
self.col.append(self.interval_attributes[i])
for i in range(self.n_binary):
self.col.append(self.binary_attributes[i])
if self.nominal_encoding==None:
for i in range(self.n_nominal):
self.col.append(self.nominal_attributes[i])
else:
for i in range(self.n_onehot):
self.col.append(self.onehot_attributes[i])
self.go_flag = True
def fit(self, df, data_map=None):
self.df_copy = deepcopy(df)
#self.df_copy = df
if data_map==None and self.features_map==None:
raise ValueError(" Call to ReplaceImputeEncode missing required"+
" Data Map.\n Use function draft_data_map to draft a map.")
sys.exit()
if type(self.features_map)==dict:
pass
elif type(data_map)==str:
try:
self.features_map = self.load_data_map(data_map)
except:
raise ValueError("Unable to load data map:", data_map)
sys.exit()
elif type(data_map)==dict:
self.features_map = data_map
else:
raise ValueError("Supplied Data Map not Dictionary or File")
sys.exit()
self.interval_attributes = []
self.nominal_attributes = []
self.binary_attributes = []
self.onehot_attributes = []
self.onehot_cats = []
self.hot_drop_list = []
for feature,v in self.features_map.items():
if v[0] not in DT.getDataTypes():
raise TypeError(
"\n***Data Map in call to ReplaceImputeEncode invalid.\n"+
"***Data Type for '"+ feature + "' is not recognized. "+
"\n***Valid types are: DT.Interval, DT.Binary, DT.Nominal, "+
"DT.Text, DT.String, DT.ID, or DT.Ignore")
if v[0]==DT.Interval:
self.interval_attributes.append(feature)
else:
if v[0]==DT.Binary:
self.binary_attributes.append(feature)
else:
if v[0]==DT.Nominal:
self.nominal_attributes.append(feature)
self.onehot_cats.append(list(v[1]))
for i in range(len(v[1])):
if type(v[1][i])==int:
my_str = feature+str(v[1][i])
else:
my_str = feature+("%i" %i)+":"+ \
str(v[1][i])[0:10]
self.onehot_attributes.append(my_str)
if self.drop==True:
self.hot_drop_list.append(my_str)
else:
if v[0] in DT.getDataTypes():
continue
else:
# Data Map Invalid
raise TypeError(
"***Data Map in call to ReplaceImputeEncode invalid.\n"+
"***Data Type for '"+ feature + "' invalid")
sys.exit()
self.n_interval = len(self.interval_attributes)
self.n_binary = len(self.binary_attributes)
self.n_nominal = len(self.nominal_attributes)
self.n_onehot = len(self.onehot_attributes)
self.cat = self.n_binary + self.n_nominal
self.n_obs = df.shape[0]
self.n_ignored = df.shape[1] - \
self.n_interval-self.n_binary-self.n_nominal
self.col = []
for i in range(self.n_interval):
self.col.append(self.interval_attributes[i])
for i in range(self.n_binary):
self.col.append(self.binary_attributes[i])
if self.nominal_encoding==None:
for i in range(self.n_nominal):
self.col.append(self.nominal_attributes[i])
else:
for i in range(self.n_onehot):
self.col.append(self.onehot_attributes[i])
if self.display:
print("\n********** Data Preprocessing ***********")
print("Features Dictionary Contains:\n%i Interval," \
%self.n_interval, "\n%i Binary," %self.n_binary,\
"\n%i Nominal, and" %self.n_nominal, \
"\n%i Excluded Attribute(s).\n" %self.n_ignored)
print("Data contains %i observations & %i columns.\n" %df.shape)
self.initial_missing = df.isnull().sum()
self.feature_names = np.array(df.columns.values)
for feature in self.feature_names:
if self.initial_missing[feature]>(self.n_obs/2):
warnings.warn(feature+":has more than 50% missing." +
"Recommend setting Data Type set to DT.Ignore.")
# Initialize number missing in attribute_map
for feature,v in self.features_map.items():
try:
self.missing_counts[feature] = self.initial_missing[feature]
except:
raise RuntimeError(feature + " is not found in Data_Map. ")
sys.exit()
# Scan for outliers among interval attributes
nan_map = df.isnull()
for index in df.iterrows():
i = index[0]
# Check for outliers in interval attributes
for feature, v in self.features_map.items():
if nan_map.loc[i,feature]==True:
continue
if v[0]==DT.Interval: # Interval Attribute
if type(v[1]) != tuple or len(v[1]) != 2:
raise ValueError("\n" +\
"***Call to ReplaceImputeEncode invalid.\n"+\
"*** Attribute Map has invalid description " +\
"for " +feature)
sys.exit()
l_limit = v[1][0]
u_limit = v[1][1]
# Check if the value is numeric before comparing
try:
value = df.loc[i,feature]
if pd.isna(value) or pd.isnull(value):
continue # Skip NaN/None values
# Try to convert to float for comparison
numeric_value = float(value)
if numeric_value > u_limit or numeric_value < l_limit:
self.outlier_counts[feature] += 1
self.df_copy.loc[i,feature] = None
except (ValueError, TypeError):
# If conversion fails, treat as outlier and convert to None
self.outlier_counts[feature] += 1
self.df_copy.loc[i,feature] = None
else:
if v[0]!=DT.Binary and v[0]!=DT.Nominal:
# don't touch this attribute
continue
# Categorical Attribute
in_cat = False
for cat in v[1]:
if df.loc[i,feature]==cat:
in_cat=True
continue
if in_cat==False:
self.df_copy.loc[i,feature] = None
self.outlier_counts[feature] += 1
if self.display:
print("\nAttribute Counts")
max_label = 0
for k, v in self.features_map.items():
if len(k) > max_label:
max_label = len(k)
max_label += 2
label_format = ("{:.<%i" %(max_label+5))+"s}{:>8s}{:>10s}"
print(label_format.format('', 'Missing', 'Outliers'))
label_format = ("{:.<%i" %max_label)+"s}{:10d}{:10d}"
for k,v in self.features_map.items():
print(label_format.format(k, self.missing_counts[k], \
self.outlier_counts[k]))
#print(label_format.format(k, v[2][0], v[2][1]))
def draft_data_map(self, df, max_n=10, max_s=30, display_map=True,
out=None, replace=False):
feature_names = np.array(df.columns.values)
draft_features_map = {}
print("\nGenerating DATA_MAP for use in ReplaceImputeEncode.")
print("String attributes with fewer than", max_s,
"unique values are labeled as Binary or Nominal; "+
"otherwise Text or String.")
print("Numerical attributes with fewer than", max_n,
"unique values are considered Binary or Nominal;"+
" otherwise Interval")
for feature in feature_names:
n = df[feature].value_counts()
n_values = n.index
astring = False
for v in n_values:
if (type(v)==str):
astring = True
break
if astring == False:
min_ = round(df[feature].min()-0.5,4)
max_ = round(df[feature].max()+0.5,4)
if astring == False:
#Numerical attribute
if len(n) < max_n:
# Numerical Attribute is Binary or Nominal
a = df[feature].unique()
# Look for string in a
j = 0
for i in range(len(a)):
if type(a[i])==str:
j += 1
if j>0:
print("WARNING: ", feature, "contains both numbers "+
"and strings. Dropping from draft data map.")
break
a.sort()
categories = tuple(a)
if len(a) == 2:
draft_features_map[feature]=[DT.Binary,
categories]
else:
draft_features_map[feature]=[DT.Nominal,
categories]
else:
# Attribute is Interval
draft_features_map[feature]=["DT.Interval",
(min_, max_)]
else:
# String Attribute is Binary, Nominal or Text or String
if len(n) < max_s:
# String Attribute is Binary or Nominal
a = df[feature].unique()
# Look for nan in a
no_nan = False
while no_nan == False:
j = -1
for i in range(len(a)):
if type(a[i]) != str:
j = i
if j>=0:
a = np.delete(a,j)
else:
no_nan=True
a.sort()
categories = tuple(a)
if len(a) == 2:
draft_features_map[feature]=[DT.Binary,
categories]
else:
draft_features_map[feature]=[DT.Nominal,
categories]
else:
k = df[feature].str.len().max()
if k>100:
# Set attribute to text field
draft_features_map[feature]=[DT.Text,("")]
else:
draft_features_map[feature]=[DT.String,("")]
if display_map:
# print the features map
print("************* DRAFT DATA MAP **************\n")
print("data_map = {")
for feature,v in draft_features_map.items():
w = DT.convertDataType(v[0])
s = "\t["
if len(feature)<5:
s = "\t\t["
print("\t'"+feature+"':",s,str(w),",",v[1],"],")
print("\n}")
if replace==True:
# Use this draft map for RIE processing
self.features_map = draft_features_map
print("Using Draft Data Map for ReplaceImputeEncode.\n"+
"Review Draft for Data Type Accuracy.")
if out != None:
#Save this draft map as a pickle file <out>
self.save_data_map(draft_features_map, out)
print(draft_features_map)
return draft_features_map
def update_feature(self, feature, datatype, dataval):
if type(feature)!=str:
raise ValueError("feature name not string")
sys.exit()
if datatype!=DT.Interval and datatype!=DT.Binary and \
datatype!=DT.Nominal and datatype!=DT.Text and \
datatype!=DT.Ignore and datatype!=DT.String:
raise ValueError("Data Type Value Invalid")
sys.exit()
if type(self.features_map)!=dict:
self.features_map = {}
self.features_map[feature] = [datatype, dataval]
def save_data_map(self, data_map, fname):
if type(data_map)!=dict:
raise RuntimeError("Data Map invalid")
sys.exit()
try:
with open(fname, 'wb') as f:
pickle.dump(data_map, f,
pickle.DEFAULT_PROTOCOL)
print("Data Map Saved to Pickle File: ", fname)
except:
warnings.warn("Cannot save data map into file: ")
def get_data_map(self):
# Returns current data map
return self.features_map
def load_data_map(self, fname):
try:
with open(fname, 'rb') as f:
data_map = pickle.load(f)
except:
raise ValueError("Unable to load data map from:", fname)
sys.exit()
if type(data_map)!=dict:
raise ValueError("Unable to load data map from:", fname)
sys.exit()
return data_map
def display_data_map(self):
# Display Data Map Dictionary
try:
if self.features_map==None:
raise RuntimeError("Data Map Does not Exist")
except:
raise RuntimeError("Data Map Does not Exist")
# print the features map
print("************* CURRENT DATA MAP **************\n")
print("data_map = {")
for feature,v in self.features_map.items():
w = DT.convertDataType(v[0])
c = ','
if w=='DT.Ignore' or w=='DT.Binary':
c = ' ,'
elif w=='DT.Nominal':
c = ' ,'
elif w=='DT.Text' or w=='DT.String':
c = ' ,'
s = "\t["
if len(feature)<5:
s = "\t\t["
print("\t'"+feature+"':",s,str(w), c, v[1],"],")
print("\n}")
def impute(self):
self.impute_interval()
self.impute_binary()
self.impute_nominal()
self.imputed_data()
def impute_interval(self):
if (self.n_interval==0):
self.imputed_interval_data = np.empty((self.n_obs, 0))
return
# Put the interval data from the dataframe into a numpy array
#depricated= self.df_copy.as_matrix(columns=self.interval_attributes)
interval_data= self.df_copy[self.interval_attributes].values
# Create the Imputer for the Interval Data
#self.interval_imputer = preprocessing.Imputer(strategy='mean')
self.interval_imputer = SimpleImputer(strategy='mean')
# Impute the missing values in the Interval data
self.imputed_interval_data = \
self.interval_imputer.fit_transform(interval_data)
def impute_binary(self):
if (self.n_binary==0):
self.imputed_binary_data = np.empty((self.n_obs, 0))
return
# Put the binary data from the dataframe into a numpy array
#cat_df = df[self.binary_attributes]
cat_df = pd.DataFrame(columns=self.binary_attributes)
for feature in self.binary_attributes:
#cat_df[feature]= self.df_copy[feature].astype('category').cat.codes
cat_df[feature]= self.df_copy[feature].astype('category')
cat_df.loc[cat_df[feature]==-1, feature] = None
#Depricated cat_array = cat_df.as_matrix()
cat_array = cat_df.values
# Create Imputer for Categorical Data
#cat_imputer = preprocessing.Imputer(strategy='most_frequent')
cat_imputer = SimpleImputer(strategy='most_frequent')
# Impute the missing values in the Categorical Data
self.imputed_binary_data = \
cat_imputer.fit_transform(cat_array)
def impute_nominal(self):
if (self.n_nominal==0):
self.imputed_nominal_data = np.empty((self.n_obs, 0))
return
# Put the nominal data from the dataframe into a numpy array
cat_df = pd.DataFrame(columns=self.nominal_attributes)
for feature in self.nominal_attributes:
#self.cat_df[feature]= self.df_copy[feature].astype('category').cat.codes
cat_df[feature]= self.df_copy[feature].astype('category')
cat_df.loc[cat_df[feature]==-1, feature] = None
#Depricated cat_array = cat_df.as_matrix()
#print(cat_array[0])
cat_array = cat_df.values
# Create Imputer for Categorical Data
#cat_imputer = preprocessing.Imputer(strategy='most_frequent')
cat_imputer = SimpleImputer(strategy='most_frequent')
# Impute the missing values in the Categorical Data
self.imputed_nominal_data = \
cat_imputer.fit_transform(cat_array)
def imputed_data(self):
# Bring Interval and Categorial Data Together into a dataframe
# The Imputed Data
# Col is not the same as self.col. col contains the main attribute
# names, self.col contains the one-hot names
col = self.interval_attributes + self.binary_attributes + \
self.nominal_attributes
# if no_impute is given, replace these attributes with their
# original, missing values
if self.no_impute != None:
idx = []
for i in range(len(self.no_impute)):
idx.append(-1)
for i in range(len(col)):
for j in range(len(self.no_impute)):
if col[i] == self.no_impute[j]:
idx[j] = i
break
for j in range(len(self.no_impute)):
k = idx[j]
if k < 0:
warnings.warn(" \nArgument "+self.no_impute[j]+ \
" in 'no_impute' is invalid.\n")
break
if k<self.n_interval:
for i in range(self.n_obs):
self.imputed_interval_data[i,k] = \
self.df_copy[self.no_impute[j]][i]
else:
if k < self.n_interval + self.n_binary:
k = k - self.n_interval
for i in range(self.n_obs):
self.imputed_binary_data[i,k] = \
self.df_copy[self.no_impute[j]][i]
else:
k = k - self.n_interval - self.n_binary
for i in range(self.n_obs):
self.imputed_nominal_data[i,k] = \
self.df_copy[self.no_impute[j]][i]
self.data_imputed= \
np.hstack((self.imputed_interval_data,\
self.imputed_binary_data, \
self.imputed_nominal_data))
self.imputed_data_df = \
pd.DataFrame(self.data_imputed, columns=col)
def scale_encode(self):
self.standardize_interval()
self.encode_binary()
self.encode_nominal()
self.encoded_data()
def standardize_interval(self):
if (self.n_interval==0 or self.interval_scale==None):
self.scaled_interval_data = self.imputed_interval_data
return
# Standardize Interval Data using Z-Scores
if self.interval_scale=='std':
scaler = preprocessing.StandardScaler()
scaler.fit(self.imputed_interval_data)
self.scaled_interval_data = \
scaler.transform(self.imputed_interval_data)
# Standardize Interval Data using median and IQR
if self.interval_scale=='robust':
scaler = preprocessing.RobustScaler()
scaler.fit(self.imputed_interval_data)
self.scaled_interval_data = \
scaler.transform(self.imputed_interval_data)
def encode_binary(self):
# Uses 1 and -1 encoding for binary instead of 0, 1
# SAS uses the 1, -1 convention
if self.n_binary == 0 or self.binary_encoding == None:
return
if self.binary_encoding == 'SAS':
low = -1
else: # One-hot encoding
low = 0
for j in range(self.n_binary):
k = self.imputed_binary_data[0:,j].argmin()
smallest = self.imputed_binary_data[k,j]
for i in range(self.n_obs):
if self.imputed_binary_data[i,j] == smallest:
self.imputed_binary_data[i,j] = low
else:
self.imputed_binary_data[i,j] = 1
def encode_nominal(self):
if (self.n_nominal==0 or self.nominal_encoding==None):
return
# Create an instance of the OneHotEncoder & Selecting Attributes
# Attributes must all be non-negative integers
# Missing values may show up as -1 values, which will cause an error
onehot = preprocessing.OneHotEncoder(categories=self.onehot_cats)
self.hot_array = \
onehot.fit_transform(self.imputed_nominal_data).toarray()
n_features = []
nominal_categories = 0
for i in range(self.n_nominal):
feature = self.nominal_attributes[i]
v = self.features_map[feature]
n_features.append(len(v[1]))
nominal_categories += len(v[1])
if nominal_categories < self.hot_array.shape[1]:
raise RuntimeError(' Call to ReplaceImputeEncode Invalid '+ \
' Number of one-hot columns is', self.hot_array.shape[1], \
'but nominal categories is ', nominal_categories, \
' Data contains more nominal attributes than '+ \
'found in the data_map.')
sys.exit()
# SAS Encoding subtracts the last one-hot vector from the others,
# for each nominal attribute.
if self.nominal_encoding == 'SAS':
self.sas_encoded = \
np.zeros((self.n_obs, (self.n_onehot-self.n_nominal)))
ilast = -1
idx1 = 0
idx2 = 0
for l in range(self.n_nominal):
m = n_features[l]
ilast = ilast + m
for j in range(m-1):
for i in range(self.n_obs):
last = self.hot_array[i,ilast]
self.sas_encoded[i,idx1] = \
self.hot_array[i,idx2] - last
idx1 += 1
idx2 += 1
idx2 += 1
def encoded_data(self):
# Bring encoded and scaled data together into a dataframe
# The Imputed and Encoded Data
if self.n_nominal==0:
if self.interval_scale==None:
self.data_encoded = np.hstack((self.imputed_interval_data, \
self.imputed_binary_data))
else:
self.data_encoded = np.hstack((self.scaled_interval_data, \
self.imputed_binary_data))
if self.drop==True:
for i in range(self.n_nominal): #WEIRD! n_nominal=0?
self.col.remove(self.hot_drop_list[i])
if self.n_nominal>0 and self.nominal_encoding==None:
if self.interval_scale==None:
self.data_encoded = np.hstack((self.imputed_interval_data, \
self.imputed_binary_data, \
self.imputed_nominal_data))
else:
self.data_encoded = np.hstack((self.scaled_interval_data, \
self.imputed_binary_data, \
self.imputed_nominal_data))
if self.n_nominal>0 and self.nominal_encoding == 'SAS':
if self.interval_scale==None:
self.data_encoded = np.hstack((self.imputed_interval_data, \
self.imputed_binary_data, \
self.sas_encoded))
else:
self.data_encoded = np.hstack((self.scaled_interval_data, \
self.imputed_binary_data, \
self.sas_encoded))
if self.drop==True:
for i in range(self.n_nominal):
self.col.remove(self.hot_drop_list[i])
if self.n_nominal>0 and self.nominal_encoding == 'one-hot':
if self.interval_scale==None:
self.data_encoded = np.hstack((self.imputed_interval_data, \
self.imputed_binary_data, \
self.hot_array))
else:
self.data_encoded = np.hstack((self.scaled_interval_data, \
self.imputed_binary_data, \
self.hot_array))
# data_encoded array ready for conversion to dataframe
self.encoded_data_df = \
pd.DataFrame(self.data_encoded, columns=self.col)
if self.nominal_encoding == 'one-hot' and self.drop==True:
self.encoded_data_df = \
self.encoded_data_df.drop(self.hot_drop_list, axis=1)
for i in range(self.n_nominal):
self.col.remove(self.hot_drop_list[i])
def transform(self):
self.impute()
self.scale_encode()
#Check for constant data columns
feature_names = np.array(self.encoded_data_df.columns.values)
for feature in feature_names:
if feature in self.interval_attributes:
self.encoded_data_df[feature] = \
self.encoded_data_df[feature].astype('float64')
elif feature in self.onehot_attributes:
self.encoded_data_df[feature] = \
self.encoded_data_df[feature].astype('int')
elif feature in self.binary_attributes:
if self.binary_encoding != None:
self.encoded_data_df[feature] = \
self.encoded_data_df[feature].astype('int')
else:
self.encoded_data_df[feature] = \
self.encoded_data_df[feature].astype(\
self.df_copy[feature].dtype)
else:
if type(self.encoded_data_df[feature]) == int:
self.encoded_data_df[feature] = \
self.encoded_data_df[feature].astype('int')
else:
self.encoded_data_df[feature] = \
self.encoded_data_df[feature].astype(\
self.df_copy[feature].dtype)
n = self.encoded_data_df[feature].value_counts()
if len(n)==1:
print("WARNING: Data for ", feature, " is constant.")
return self.encoded_data_df
def fit_transform(self, df, data_map=None):
self.fit(df, data_map)
self.transform()
return self.encoded_data_df
|