-
Notifications
You must be signed in to change notification settings - Fork 0
/
README
49 lines (34 loc) · 1.87 KB
/
README
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
# Predictive Regression Model using Scikit-Learn and PyCaret
## Project Overview
This project develops a predictive regression model to estimate the "Selling_Price" based on different factors. We leverage the power of scikit-learn and PyCaret, two of the most popular machine learning libraries in Python, to preprocess data, select features, train models, and evaluate performance.
### Objectives
- To preprocess and analyze the dataset for regression analysis.
- To select the most significant features affecting the "Selling_Price".
- To compare multiple regression models and select the best performer based on different metrics (MSE, RMSE, etc).
- To fine-tune the model parameters for optimal performance.
- etc.
## Dataset
The dataset used for our regression models is public and can be found in the Kaggle datasets section (https://www.kaggle.com/datasets/nehalbirla/vehicle-dataset-from-cardekho).
### Features
- **Age**: [Age of the vehicule]
- **Present_Price**: [Present price for the vehicule]
- **Kms_Driven**: [Kilometers driven using the car]
- **Fuel_Type**: [Fuel type : Diesel or Petrol or CNG]
- **Seller_Type**: [Seller type : Dealer or individual]
- **Transmission**: [Transmission : Manual or Automatic]
- **Owner**: [Owner of the car]
### Target Variable
- **[Selling_Price]**: [Description]
## Installation
This project is executed using Python 3.11 and the following Python libraries:
- NumPy
- Pandas
- scikit-learn
- PyCaret
- matplotlib (for data visualization)
- seaborn (for data visualization)
- etc.
To run the jupyter notebook in a container, follow the steps below:
- build the docker image using the provided dockerfile by running the command "docker build -t regression ."
- run a container using the image you just created by running the command "docker run -p 8888:8888 regression"
- to access your virtual environment, use the link obtained via the command line.