Scribd
Upload
3 views22 pages

ML_intro

The document provides an overview of machine learning, including its definition as a subset of AI that enables systems to learn from data and make decisions. It categorizes machine learning into supervised and unsupervised types, detailing their respective algorithms and applications. Additionally, it lists essential libraries and tools for machine learning and outlines the structure of the next class focusing on datasets and relevant programming libraries.

Uploaded by

bibekchataut123
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
3 views22 pages

ML_intro

The document provides an overview of machine learning, including its definition as a subset of AI that enables systems to learn from data and make decisions. It categorizes machine learning into supervised and unsupervised types, detailing their respective algorithms and applications. Additionally, it lists essential libraries and tools for machine learning and outlines the structure of the next class focusing on datasets and relevant programming libraries.

Uploaded by

bibekchataut123
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/ 22

Machine Learning

Gagan Puri
linkedin.com/in/puri-gagan
[email protected]
Contents
●
AI Introduction
●
Machine Learning Types
●
Essential Libraries and Tools
●
Anaconda and Environment Management
●
How it looks like
●
Next Class
Overwhelmingness
Structure
●
Real quick revision of previous class
●
Pre-Cap
●
Short discussion on the DYOR topic if given previously
●
Proceed ….
●
Interactive Examples with Jupyter Notebook
●
Over: with short DYOR
Introduction

●
AI is the broadest term of the four. It refers to the development of machines that can perform
tasks that typically require human intelligence
●
AI is about creating “smart” machines that can mimic human behaviors, make decisions, solve
problems, understand language
●
Autonomous Systems

●
ML is a subset of AI that focuses on creating systems that
can learn from data, identify patterns, and make decisions
without explicit programming
●
Once trained, the model can make predictions or decisions
based on new, unseen data
●
DL is a specialized branch of ML that focuses on neural
networks algorithm that mimic the way the human brain
processes information
Still… What’s difference? ML vs DL

●
How ML works:
●
Training: You feed a model with historical data (sample data)
●
Learning: The model identifies patterns in the data
●
Prediction: The model makes predictions about new data
●
How Deep Learning works:
●
Input Layer: Raw data, such as an image, is fed into the system
●
Hidden Layers: Multiple layers of neurons process the data by applying mathematical
transformations
●
Output Layer: The network generates a prediction, such as Classification: e.g. whether a
photo contains a cat or a dog
And what about Data Science
●
Data Science is an interdisciplinary field that focuses on extracting insights from large datasets
●
Use a combination of statistics, programming, machine learning, and domain knowledge
●
Analyze data and drive decision-making

Why it’s certain domain is outside of AI:


●
Data Science is not about creating intelligent systems
●
It’s about understanding and interpreting data to
inform business strategies, product development, and
more
●
Data analysis, statistics, and domain expertise to
derive actionable insights from data
●
Can use AI
Machine Learning

●
Teaches computers to recognize patterns and make decisions automatically using data and
algorithms

●
Broadly categorized into three types
Types
Supervised Machine Learning
●
A model is trained on labeled data: each input is paired with the correct output
●
Learning a class from examples: providing the model with examples where the correct label is known
Supervised Machine Learning Types
●
Applied to two main types of problems
●
Classification: Where the output is a categorical variable (e.g., spam vs. non-spam emails, yes vs. no)
●
Regression: Where the output is a continuous variable (e.g., predicting house prices, stock prices)
Supervised
Machine Learning
Algorithms
Unsupervised Machine Learning
●
Tasked with finding patterns and relationships within the data without any prior knowledge of the data’s
meaning
●
The training model has only input parameter values and discovers the groups or patterns on its own
●
Finds hidden patterns and data without any human intervention: we don’t give output to our model
Unsupervised Machine Learning Types
●
Applied to 3 main types of Algorithms
●
Clustering
●
K-means Clustering: Groups data into K clusters based on how close the points are to each other
●
Association Rule Learning
●
Apriori Algorithm: Finds patterns by exploring frequent item combinations step-by-step
●
Dimensionality Reduction
●
PCA: Reduces dimensions by transforming data into uncorrelated principal components.
Example Application: Unsupervised Machine Learning
●
You have 1,000 news articles from various sources, covering a range of topics like politics, sports,
technology, health, and entertainment without any manual labeling. After clustering:
●
Group 1: Articles about the latest political developments (e.g., "Senator proposes new healthcare
bill," "President addresses climate change issues").
●
Group 2: Articles about football and basketball games (e.g., "Champions League final preview,"
"NBA playoffs predictions").
●
Group 3: Articles on advancements in AI and new tech gadgets (e.g., "AI models surpass human-
level performance," "New smartphone launches this year").
Essential Libraries and Tools
●
Conda
●
Jupyter Notebook
●
NumPy
●
Pandas
●
Matplotlib / Seaborn
●
Sklearn
●
TensorFlow
●
PyTorch
●
XGBoost & LightGBM
●
SciPy
●
NLTK / Spacy
●
OpenCV
Installation
Anaconda / Mini Conda Environments
Managing Environments
Managing Environments
Codes?
Next Class
●
Datasets
●
Numpy
●
Scikit Learn
●
Matplotlib
●
Pandas

You might also like