Auto Data cleaner - “samoy”
===========================

**samoy** is a Python package for machine learning and data science,
built on top of Pandas inbuilt libraries. This package will be useful
for data pre-processing before starting off any machine learning or data
science project as it will ease your entire process of data cleaning
without much input from the user. This project is developed in July,2020
by a team of post graduate students and since then many have also
contributed. It is currently maintained by the same team. ##
Installation Dependencies

samoy requires:

-  Python (>= 3.6)
-  NumPy (>= 1.13.3)

 User installation

If you already have a working installation of numpy and pandas, the
easiest way to install samoy is using ``pip``

::

   pip install -U samoy

or ``conda``:

::

   conda install -c conda-forge samoy

Package Utilities
-----------------

In current version of this package it only handles missing,null values
and duplicates along with case conversion

Utilities supported:

-  Null values handeling : Dropping and imputing(with mean,median as
   well as custom values)
-  Imputing null values with the threshold as mentioned by the user in
   terms of percentage
-  Missing values handeling : Dropping and imputing(with mean,median and
   LRU)
-  Duplicates removal
-  Case conversion like if use wants to convert specific columns or
   entire column values into either lower or upper case

Name of functions provided in this package: 1. Handeling Null - dropnull
: This function will drop null in three ways that is dropping all null
in the entire dataframe,dropping columns or rows having all nulls and
dropping the rows or columns having any of the value as null and return
the dataframe after removing null by the method as mentioned by user. By
default it drops all the null if no method is mentioned explicitely. -
dropnull_th : This function will drop the nulls in those columns where
the number of nulls is greater than or equal to the percentage specified
by the user and returns dataframe having nulls dropped in those columns
where number of null is greater than the percentage(mentioned by user
while calling function) of total number of records in that column. -
swapnull : This function will replace all the null values with the three
different methods like custom method,mean and median and it will return
the dataframe having all null values replaced by the method as chosen by
the user.By default if nothing is mentioned explicitely,it will impute
nulls with the mean value. - swapnull_subset : This function will
replace all the null values in the columns as specified by the user with
the three different methods like custom method,mean and median and it
will return the dataframe having all null values replaced in the
selected columns by the method as chosen by the user.By default if
nothing is mentioned explicitely,it will impute nulls with the mean
value. 2. Handeling Missing - swapmissing : This function replace NaN
values with mean or median of the specific column. If user mention
method as mean or median it will pick it as it is and if user dont want
to mention any method then by default mean value to be replaced. -
swapmissing_subset : This function replace NaN values with mean or
median in the specific **numeric** columns only. If user mentions any
method like mean or median it will pick it as it is and if user dont
want to mention any method then by default missing value will be
replaced to mean value. - swapmissing_lru : This function will replace
the NaN value with last and next value of the same column and if there
are many NaN values with start of the column then it will start
replacing same with mean of the same column. - dropmissing : This
function drops missing values from the given data. - dropmissing_rows :
This function drops rows whose all values are missing. -
dropmissing_subset : This function drops missing values as well as NaN
values from the selcted columns of the dataframe 3. Handeling Duplicates
- drop_replicatecols : This function will drop duplicate columns and
return the dataframe after dropping the columns having same name(names
are also typographically matched with respect to upper or lower case). -
drop_replicates : This function drops duplicate rows that are present in
enitre column of dataframe or within the subset of columns as specifies
by users. 4. Case conversion - altercase : This function is usually to
convert the content of dataframe either in lower case or upper case. -
altercase_subset : This function is usually to convert the content of
only selected columns either in lower case or upper case and returns
dataframe having the content of mentioned columns either in lower or
upper case.

Details of each functions

To know the detailed summary of each function along with its parameters
and a demo example to use it,please use command
``help(name_of_function)``

Contact Us:
-----------

Email: samoyapi@gmail.com
