first commit

Author: singhrahuldps

LICENSE.txt

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2018 Rahul
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

OneHotEncode/OneHotEncode.py

@@ -0,0 +1,81 @@
+def OneHotEncode(pandas_dataframe,category_columns=[],check_numerical=False,max_var=None):
+	# Parameter explanation ----
+	# pandas_dataframe -> The Pandas Dataframe object that contains the column you want to one-hot encode
+	# category_columns -> List of column names in pandas_dataframe that you want to one-hot encode
+	# override_alert (Default=False) -> A naive way of checking if the column contains numerical
+	#                                   data or is unsuitable for one-hot encoding
+	#                                   Set it to True to turn on the detection
+
+	import numpy as numpylib
+
+	# The dataframe is copied to a new variable
+	df=pandas_dataframe.copy()
+
+	# List of list of names of all new columns made for a single column 
+	all_new_cols=[]
+
+	# List of dictionary of names of all new columns made for a single column
+	new_col_dict=[]
+
+	# List of arrays containg the dropped columns that were originally input
+	dropped_cols=[]
+
+	numerical_const=20
+	
+	for col in category_columns:
+
+		# Total number of rows in each column
+		total_rows=len(df[category_columns[0]].values)
+
+		category_elements=[]
+		main_row=df[str(col)].values
+		for category_element in main_row:
+			category_elements.append(category_element)
+		category_elements=list(set(category_elements))
+
+		if check_numerical:
+			if max_var!=None:
+				if len(category_elements) > max_var:
+					print(col+' not suitable for One-Hot Encoding')
+					continue
+			else:
+				if len(category_elements)>numerical_const:
+					print(col+' has more variables than permitted')
+					continue
+		
+		if max_var != None:
+			if len(category_elements) > max_var:
+					print(col+' has more variables than allowed')
+					continue
+
+		category_element_dict={}
+		for i in range(len(category_elements)):
+			category_element_dict[category_elements[i]]=i
+
+		category_element_reverse_dict={}
+		for i in range(len(category_elements)):
+			category_element_reverse_dict[col+str(i)]=category_elements[i]
+
+		dict_new_columns={}
+
+		category_element_str = [(col+str(x)) for x in range(len(category_elements))]
+		for string in category_element_str:
+			zero_row=numpylib.zeros((total_rows,), dtype=numpylib.int)
+			dict_new_columns[string]=zero_row
+
+		colnames=[]
+		for i in range(total_rows):
+			colnames.append(category_element_str[category_element_dict[main_row[i]]])
+		for i in range(total_rows):
+			dict_new_columns[colnames[i]][i]=1
+		for element in category_element_str:
+			df[element]=dict_new_columns[element]
+
+		# Original columns are dropped from the dataframe
+		df=df.drop(col,1)
+
+		all_new_cols.append(category_element_str)
+		new_col_dict.append(category_element_reverse_dict)
+		dropped_cols.append(main_row)
+
+	return df,dropped_cols,all_new_cols,new_col_dict

OneHotEncode/__init__.py

@@ -0,0 +1,21 @@
+# One-Hot Encode
+A python script to deploy One-Hot encoding in Pandas Dataframes
+
+Requirements -> Pandas, Numpy
+
+#### Input -> (pandas_dataframe,cols,check_numerical=False,max_var=20)
+	pandas_dataframe -> The Pandas Dataframe object that contains the column you want to one-hot encode	 
+	cols -> List of column names in pandas_dataframe that you want to one-hot encode
+	check_numerical (Default=False) -> A naive way of checking if the column contains numerical
+	                                   data or is unsuitable for one-hot encoding
+	                                   Set it to True to turn on the detection
+	max_var (Default=20) -> Max number of diferent variables allowed in a category
+                                     
+#### Returns df,dropped_cols,all_new_cols,new_col_dict
+	df -> Pandas dataframe that returns the one-hot encoded data with the original columns dropped
+	dropped_cols -> List of arrays containg the dropped columns that were originally input
+	all_new_cols -> List of list of names of all new columns made for a past single column 
+	new_col_dict -> List of dictionary of names of all new columns made for a past single column
+   
+
+More info would be soon provided through a Medium page.

README.md

@@ -0,0 +1,2 @@
+[metadata]
+description-file = README.md

setup.cfg

@@ -0,0 +1,13 @@
+from distutils.core import setup
+setup(
+  name = 'OneHotEncode',
+  packages = ['OneHotEncode'], # this must be the same as the name above
+  version = '0.1',
+  description = 'A python script to deploy One-Hot encoding in Pandas Dataframes',
+  author = 'Rahul Singh',
+  author_email = 'singhrahuldps@gmail.com',
+  url = 'https://github.com/singhrahuldps/OneHotEncode.git', # use the URL to the github repo
+  download_url = 'https://github.com/singhrahuldps/OneHotEncode/archive/0.1.tar.gz', # I'll explain this in a second
+  keywords = ['pandas', 'database', 'one-hot','one','hot','features','categorical','numerical','encode','encoding'], # arbitrary keywords
+  classifiers = [],
+)