Classification Laboratory 📉

The Classification Laboratory is a dynamic model that accepts csv data from users for classification issues with a binary outcome as the dependent variable.

This laboratory assists users in gaining insight into data, and users will receive an EDA report📑 based on pandas profiling of the file they provide.

A classifier manager aids in the management of various machine learning algorithms and the selection of the most appropriate parameters for all models.Deploy a classification laboratory using streamlit

.

What you will explore in this classification laboratory

• EDA report

• Train Test split 

• Feature Engineering

    - PCA 

• Machine Learining models

  - K Nearest neighbour

   - Logistic Regression

   - Decision Tree

   - Random Forest Classifier

   - AdaBoostClassifier

   - GradientBoostingclassifier

• You will obtain the following outcomes:👇

  -Variance bias tradoff

  -Classification Report

  -Confusion Metrix

  -Mcnemar test 

  -ROC Curve 

  -Best Parameter

Demo

Installation

You need these dependencies:

pip install streamlit

Pandas profiling

pip install pandas-profiling

To install mlxtend, just execute

pip install mlxtend  

plotly 5.7.0

pip install plotly

website: https://classification-laboratory-app.herokuapp.com/