localizeflow[bot]
3b2098a6fc
|
3 months ago | |
|---|---|---|
| .. | ||
| README.md | 3 months ago | |
| assignment.md | 7 months ago | |
README.md
Introduction to machine learning
Pre-lecture quiz
🎥 Click di image wey dey top for short video wey dey run through dis lesson.
Welcome to dis course on classical machine learning for beginners! Whether you dey totally new to dis topic, or you be experienced ML practitioner wey wan brush up for one area, we dey happy to get you join us! We want create one friendly launching spot for your ML study and we go happy to check, respond, and put your feedback inside.
🎥 Click di image wey dey top for video: MIT's John Guttag dey introduce machine learning
Getting started with machine learning
Before you start with dis curriculum, you need make your computer set up well and dey ready to run notebooks locally.
- Configure your machine with these videos. Use di links dem below to learn how to install Python for your system and setup a text editor for development.
- Learn Python. E good make you get basic understanding of Python, one programming language wey data scientists dey use and we dey use for dis course.
- Learn Node.js and JavaScript. We still go use JavaScript small for dis course when we dey build web apps, so you go need to get node and npm install, plus Visual Studio Code ready for both Python and JavaScript development.
- Create a GitHub account. Since you find us here for GitHub, you fit don get account, but if no be so, make you create one then fork dis curriculum make you fit use am for your own. (Feel free make you give us star too 😊)
- Explore Scikit-learn. Make you sabi Scikit-learn, one set of ML libraries wey we dey talk about for these lessons.
Wetin be machine learning?
Di term 'machine learning' na one of di most popular and well-used words today. E get chance say you don hear dis term at least one time if you get any kinda knowledge about technology, no matter which area you dey work. But di way machine learning dey work na mystery to most people. For person wey dey start for machine learning, di subject fit hard sometimes. So e good make you understand wetin machine learning be for real, and make you learn am small-small by step, through practical examples.
The hype curve
Google Trends dey show di recent 'hype curve' of di term 'machine learning'
A mysterious universe
We dey live for one universe wey full of plenty mysteries wey dey amaze us. Great scientists like Stephen Hawking, Albert Einstein, and many more don spend dia life dey find important information wey fit uncover di mysteries of di world wey dey round us. Na so human beings be to learn: pikin dey learn new tins and dey uncover how their world be year by year as dem dey grow.
The child's brain
Pikin brain and sense dem dey gather facts about wetin dey around and dem dey learn di hidden patterns of life slowly wey go help di pikin create logical rules to sabi di patterns wey dem don learn. How human brain dey learn na wetin make humans be di most sabi animal for dis world. We dey learn steady by finding hidden patterns then we dey create new tins based on those patterns and dat dey help us better as we dey live. Dis ability to learn and change na wetin we dey call brain plasticity. For example, we fit see small similarity between how human brain dey learn and how machine learning dey work.
The human brain
Di human brain dey collect tins from di real world, dey process wetin e gather, dey make smart decisions, then dey do actions based on situation. Na wetin we dey call intelligent behavior. When we program machine make e mimic dis intelligent behavior, na wetin dem dey call artificial intelligence (AI).
Some terminology
Even though di terms fit confuse, machine learning (ML) na important part of artificial intelligence. ML dey use special algorithms to find important information and hidden patterns from data to support smart decision making.
AI, ML, Deep Learning
One diagram wey show how AI, ML, deep learning, and data science dey relate. Infographic by Jen Looper inspired by this graphic
Concepts to cover
For dis curriculum, we go only cover important machine learning concepts wey beginners need sabi. We go mainly talk about 'classical machine learning' we dey use Scikit-learn, one correct library wey many students dey use learn basics. To understand bigger concepts for artificial intelligence or deep learning, you need get correct basic knowledge for machine learning, so we wan give am here.
In this course you will learn:
- core concepts of machine learning
- history of ML
- ML and fairness
- regression ML techniques
- classification ML techniques
- clustering ML techniques
- natural language processing ML techniques
- time series forecasting ML techniques
- reinforcement learning
- real-world applications for ML
Wetin we no go cover
- deep learning
- neural networks
- AI
To make learning better, we go avoid di complicated things about neural networks, 'deep learning' - wey na many-layer model building with neural networks - and AI, we go talk about those for another curriculum. We go still bring one data science curriculum to focus on that big part of dis field.
Why study machine learning?
Machine learning from system side na to create automated systems wey fit learn hidden patterns from data to help make smart decisions.
Dis idea come from how human brain dey learn tins based on data e gather from outside world.
✅ Think small about why business go want use machine learning strategies instead of to create hard coded rules-based engine.
Why data quality matters
Good quality data go make model work well. Bad or noisy data fit make wrong predictions even if you use better machine learning algorithms.
Applications of machine learning
Machine learning dey everywhere now, e dey everywhere like data wey dey flow for our societies wey come from our smart phones, connected devices, and other systems. Because machine learning algorithms get serious power, researchers don dey explore how dem fit solve wahala wey get many sides and many fields with good results.
Examples of applied ML
You fit use machine learning for many ways:
- To predict if person get disease from their medical history or reports.
- To use weather data predict weather changes.
- To understand how text dey express feelings.
- To find fake news and stop propaganda.
Finance, economics, earth science, space exploration, biomedical engineering, cognitive science, and even humanities fields don use machine learning to solve serious, heavy data problems for their area.
Conclusion
Machine learning dey automate how e dey find patterns by getting important insights from real or generated data. E don prove say e get big value for business, health, and financial applications and others.
Soon, knowing di basics of machine learning go be must for people from any area because e don spread everywhere.
🚀 Challenge
Sketch, for paper or use online app like Excalidraw, how you understand di difference between AI, ML, deep learning, and data science. Add some ideas of problems wey each of these techniques fit solve well.
Post-lecture quiz
Review & Self Study
To learn more on how you fit work with ML algorithms for cloud, follow dis Learning Path.
Take one Learning Path about basics of ML.
Assignment
Disclaimer: Dis document don translate wit AI translation service Co-op Translator. Even tho we dey try make am correct, abeg make you know say automated translation fit get errors or mistakes. Di original document for dia own language na im be di correct source. For important info, make person wey sabi human translation do am. We no go responsible for any misunderstanding or wrong understanding wey fit happen because of dis translation.