Skip to main content

Skills application

Instruction and application
Complete

The role of an ML Analyst​

Illustration

In this skills application, you will be working with a Biochemical firm to identify the optimal relationship between a substrate concentration and its product yield, using regularization techniques and hyperparameter tuning.

Context​

BrownRock Chemicals is a biochemical firm looking to understand how a substrate concentration (the independent variable) is converted into a product yield (dependent variable) with the help of an enzyme as a catalyst.

You have been asked to use laboratory data to identify the relationship between the substrate concentration and product yield to help the company find the optimal amount to maximize profits.

The tutorial will guide you through how to use Lasso, Ridge and Elastic Net regression as well as how to use Grid Search to fine tune the hyperparameters.

At the end, you will use these skills to identify the optimal relationship in the data for your report.

Success criteria​

To successfully complete this skills application, you must submit a written report (approximately 1200-1800 words) addressing the following key areas:

  • The optimal relationship between the substrate concentration and product yield: Find the optimal relationship between the chemical substrate concentration and product yield that the company can use to determine the best amount to maximize profits.
  • Justify your decision: Explain how you found this relationship, including what model and hyperparameters you selected.
  • Explain the performance of the model: How can you be confident in the model’s output? How have you optimized the model and sought to minimize overfitting?

Instructions and materials​

Completing this activity will “unlock” the solution example on the following page.

Complete the tutorial​

Work through the Jupyter Notebook to learn how to use each regression model and fine-tune the hyperparameters. Note there will be times where you will need to update the code yourself when you see “###” in the code.

Identify the optimal formula​

Use the skills you have learned to find the optimal relationship between substrate concentration and product yield.

Write your report​

Write your report, justifying your decisions in terms of model performance.