Python pandas profiling: Why data profiling is important for you?

If you are a python pandas library user and have experience processing datasets using pandas you must know that in order to collect input dataset-specific details you may need to run lots of commands to collect various dataset information and then combine it all together.

So what if there is a python library which can provide the detailed report in a visually appealing and interactive format with just a few lines of code, wouldn’t you be excited? Indeed, absolutely.

In this video, I will be giving you an in-depth walkthrough to the pandas-profiling library which you can use with pandas to generate details report about your dataset which includes maximum details about each column individually in your dataset along with interactions, correlations, histogram, missing and duplicate values, and lots of other very useful details, all in a fully interactive and exportable HTML report format.

Content Timeline:
---------------------------
- (00:00) Video Start
- (00:07) Video Content Intro
- (01:56) Code & Jupyter Notebook Introduction
- (02:49) Library Installation
- (03:39) Pandas-profiling Library Introduction
- (07:29) Demo with Titanic Dataset
- (13:36) Demo with Titanic Dataset - Interactions
- (14:50) Demo with Titanic Dataset - Correlations
- (15:17) Demo with Titanic Dataset - Missing Values
- (17:02) Saving HTML Report
- (18:43) Profiling large dataset
- (19:51) Minimal Profiling Reporting Setting
- (20:41) Demo with Titanic Dataset - Config Metadata
- (21:36) Demo with Titanic Dataset - Config Details
- (22:37) Demo with Titanic Dataset - Config Param
- (23:17) Demo with Titanic Dataset - View Widgets
- (25:46) Demo with Titanic Dataset - Histogram Config
- (27:19) Streamlit Application with Profiling Report
- (31:31) Recap
- (32:16) Credits

GitHub URL for the samples in the Video:

Prodramp LLC

Content Creator:

Tags:
#ai #aicloud #h2oai #driverlessai #machinelearning #cloud #mlops #model #collaboration #deeplearning #modelserving #modeldeployment #keras #tensorflow #pytorch #datarobot #datahub #aiplatform #aicloud #cometml #modelmonitoring #drift #modelregistry #modelmanagement #pandas #pandasprofiling