add python script to Qlick sense

[asma-sm]

Hello,

I'm new to Qlick sens, I'm trying to add a python script to Qlick sense.

i want to make an app that allows to impute missing data to xlsx file using sickit learn imputers and then make data visualization , my python code looks  like:

# features selection then,
imputer=  IterativeImputer(random_state=0, estimator= 
                         ExtraTreesRegressor(n_estimators=10, random_state=0))
imputed_data=imputer.fit_transform(dataset_to_impute)

 i have seen these examples:

https://github.com/qlik-oss/server-side-extension/tree/master/examples/python

but i didn't know how to do it in my case!

Please, can anyone help?

[Varun1]

Hi,

I have the same query.

@sunny_talwar  can you please help us out.

Regards,

Varun

[asma-sm · Accepted solution]

Hello, I'm back again!

So, to integrate python script to Qlick Sense you can use server-side extension (SSE) it exposes  Qlik data to an external scripting/programming language (python in our case) through remote procedure calls (RPCs).

Here we are going to passe DATA to python in a request (it is which is an iterator object made of multiple bundled rows)

1. install these packages: grpcio, pandas and scikit-learn
   pip install grpcio
   pip install pandas
   python -m pip install grpcio-tools
   pip install scikit-learn
2. Edit your Settings.ini file, in windows you can find it in :
   C:\Users\nameOfUser\Documents\Qlik\Sense
3. Add the following line to the Settings files,                                                                                                                                   SSEPlugin = Imputer,localhost:50051                                                                                                                                                         (the port's number is in "__main.py" line 242,you can change it)
4.  Use the files in my git repo:                                                      https://github.com/asmaalegat/Qlick_data_imputation_example                                                                                                     (I reused files from https://github.com/danielpilla/qlik-python-sse-nfl-playbyplay-predictions )
5. Add the the application (qvf file) to the directory of qlick sense apps
6.  Close qlick sense and run   python script in cmd:
   python __main__.py
7. launch Qlick Sense app

To add other functions or customize your application according to your need you can edit
the two files:

• "functions.json" (it defines the metadata associated with each custom function )
  
  
• "__main__.py" you can edit the function "impute" or add other functions (don't forget to
  add it to the json file)

Custom functions  are callable from the Qlik Sense front-end.
In the example I made i called the function "impute" to recuperate my data with imputed data by adding
an expression in your Qlick app :
pluginName.funtionName(col1,col2,col_n)   // your function's parameters
in the example expression is added to "imputed data" table's columns:
Imputer.Impute(Age,High,Weight)

To understand well how the SSE works and each function in python's code you can read this article
https://www.axisgroup.com/data-industry-insights-blog/qlik-sense-server-side-extensions-part-13-arch...