lurkingryuu / mlbio_assignment2

Running the code

Note: Pdf file is well formatted and easy to read. Section 1.9 contains all the inferences and comparisons of the models. It is recommended to read this section first.

Requirements

• Jupyter Notebook Support
• Run pip install -r requirements.txt to install all the dependencies in your environment.

Running the code

• Run jupyter notebook in your terminal to open the notebook in your browser.
• Open the Assignment2.ipynb file and run the cells in the notebook.

Simply want to see the results?

• Open the Assignment2.pdf file to see the results of the notebook.

Author

Name	Roll Number
Yelisetty Karthikeya S M	21CS30060

Github: lurkingryuu

Assignment #2

Objective:

This assignment focuses on understanding and comparing the performance of different machine learning models for predicting cancer types based on a given dataset. You will analyse Support Vector Machines (SVM), Random forest (RF), neural network (NN) regression and other relevant techniques.

Instructions:

Code Overview:

Write a python script for the topics mentioned above Identify the main sections for data import, manipulation, model training, and result storage.

Data Description:

Briefly introduce the dataset, its columns, and the target variable (cancer types).

Discuss the SVM analysis:

Explain SVM's role in cancer type prediction. Use different kernels

Elaborate on the neural network regression analysis:

Describe the significance of neural networks in cancer type prediction. Explain the grid search process for neural network parameters. Compare neural network results with other models.

Comparison:

Contrast the performance of the three models: Discuss strengths and weaknesses of SVM, and neural networks for predicting cancer types. Identify the most suitable model for the task based on accuracy and efficiency.

Discussion:

Reflect on the broader implications of accurate cancer type prediction. Consider real-world applications of the models' performance.

Conclusion:

Summarise findings:

• Highlight the best-performing model and rationale.
• Stress the importance of thoughtful model selection and parameter tuning in machine learning.

Submission:

Create a concise document with the outlined sections. Include necessary code snippets or visuals to support explanations. Provide references for external resources used.

Link 1

Link 2

Link 3

Create a folder with the name Asgn2_. Copy your code and all your supporting files including one README file on how to execute the code. ZIP the folder and upload in the Moodle server within the deadline.

TA: Soumyadeep Bhaduri [Email: [email protected]]