Ever confused on what Machine Learning (ML) is about, and what you should do as a student to learn more about Data Science? In the second half of the session, Traci shared about the steps required in a ML Project.
Feel free to take a look at some of the things that he has shared!
Stage 1: Preliminary
Form a problem and find data depending on the domain (e.g. Time Series, Tabular). Check if data cleaning is required, does the data exist already, is scraping required, and are labels required, these are just some examples of questions.
Github is required to start a ML Project, you can go online to learn more about how to use Github.
Stage 2: Essential
Read more about existing projects and what are some of the good pointers that could be applied to your own project. Conduct Exploratory Data Analysis (EDA) to better understand how the data works.
Train a baseline model with basic ML Algorithms, and over time, train a better model that has a better performance. Do setup a README file as well to attract more people to read more about your project.
Stage 3: Advanced
Unit Testing has to be done to test the different units and functions to ensure that all the functions work as intended.
Deployment as a micro-service can be done to your own computer.