Scribd
Upload
48 views

ML - Project

This document provides instructions for a machine learning project to predict the price of high quality backpacks using a dataset of backpacks. The project requires exploring the data, building classification, regression, and ensemble models, and evaluating the models using various metrics. Submissions should include a Jupyter notebook with code, explanations of choices, and a PDF report.

Uploaded by

Aditya Darji
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
48 views

ML - Project

This document provides instructions for a machine learning project to predict the price of high quality backpacks using a dataset of backpacks. The project requires exploring the data, building classification, regression, and ensemble models, and evaluating the models using various metrics. Submissions should include a Jupyter notebook with code, explanations of choices, and a PDF report.

Uploaded by

Aditya Darji
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Price Prediction of USA High Quality Backpacks

Weightage: 15 Marks

General Instructions:
1. Organize your code in separate sections for each task. Add comments to make the code readable.
2. Notebooks without output shall not be considered for evaluation.
3. For a given dataset, where class labels may not be found, choose the right target variable and
discretize the same for classification.

Submission guidelines:
1. Upload your submission documents in .ipynb and .pdf format

Dataset:
Attached Data Set (.csv file)

EDA
1. Perform Exploratory Data Analysis to gather insight from the dataset. Write your inference about
the analysis learned from visualizations (minimum 3) [3]

Classification
CLASSIFICATION (any of the Logistic Regression / SVM / Decision Tree/ Naïve Bayes/KNN/ANN). Justify
your design choices at each step: Write as a markdown cell in jupyter notebook at the beginning of each
subsection.
1. Perform and explain necessary pre-processing / feature engineering on this dataset [0.5]
2. Perform the Machine Learning activity. Explain the choice of target attribute, classification type,
model selected with reason [1.5]
3. Quantify and explain the quality of your ML model. Explain the choice of evaluation metric [1.5]
4. Your observation about the results (Hint: comment on the problem statement and conclude the
effectiveness of the machine learning activity) [0.5]

Regression
Any of the Linear Regression (any of Gradient / Stochastic /MiniBatch)/linear basis models/KNN/Locally
weighted regression/ any of the regularization techniques). Justify your design choices at each step: Write
as a markdown cell in jupyter notebook at the beginning of each subsection.
1. Perform and explain necessary pre-processing / feature engineering on this dataset [0.5]
2. Perform the Machine Learning activity. Explain Attributes of interest, Regularization type with
reason, model selected with reason [1.5]
3. Quantify and explain the quality of your ML model. Explain the choice of evaluation metric [1.5]
4. Your observation about the results (Hint: comment on the problem statement and conclude the
effectiveness of the machine learning activity) [0.5]

Ensemble ML
Justify your design choices at each step: Write as a markdown cell in jupyter notebook at the beginning
of each subsection.
1. Perform and explain necessary pre-processing / feature engineering on this dataset [0.5]
2. Perform the Machine Learning activity. Explain Attributes of interest, base classifier chosen with
reason; model selected with reason [1.5]
3. Quantify and explain the quality of your ML model. Explain the choice of evaluation metric [1.5]
4. Your observation about the results (Hint: comment on the problem statement and conclude the
effectiveness of the machine learning activity) [0.5]

You might also like