https://github.com/watashiwasid/machine-learning-docs
Materials I collected and or prepared while studying Machine Learning in 6th semester.
https://github.com/watashiwasid/machine-learning-docs
artificial-intelligence deep-learning machine-learning neural-networks pandas scikit-learn
Last synced: 2 months ago
JSON representation
Materials I collected and or prepared while studying Machine Learning in 6th semester.
- Host: GitHub
- URL: https://github.com/watashiwasid/machine-learning-docs
- Owner: WatashiwaSid
- Created: 2025-03-05T13:55:56.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-03-05T15:53:33.000Z (over 1 year ago)
- Last Synced: 2025-03-05T16:38:08.848Z (over 1 year ago)
- Topics: artificial-intelligence, deep-learning, machine-learning, neural-networks, pandas, scikit-learn
- Language: Python
- Homepage: https://www.linkedin.com/in/comicsid
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# INTRODUCTION TO AI/ML
- ML is a subset of AI.
- AI is the meta domain that encompasses several specialized domains of learning designed to address specific tasks and challenges.
- This includes Machine Learning, Deep Learning, Generative AI, Computer Vision, Expert Systems, Neural Networks, Natural Language Processing, Robotics and more.
- The focus of study in this repo is Machine Learning, and Deep Learning to some extent.
# INTRODUCTION TO ML
- Machine Learning is a subset of Artificial Intelligence that involves studying and developing statistical models and algorithms that enable machines to make decisions based on trained data.
- At a broad level, ML algorithms can be classified into three categories:
- Supervised Learning
- Unsupervised Learning
- Reinforcement Learning
### Supervised Learning
- Supervised Learning deals with problems when the problem statement clearly defines what needs to be done. (predicted)
- Supervised models learn with labelled data.
- The variable or outcome that needs to be predicted is known as the **predictor** **variable** or **dependent variable** **(y)**.
- The other fields in the dataset are called **features** or **independent variables** **(x)**.
- Applications of Supervised Learning Algorithms:
- Customer Churn Prediction (Classification)
- Bank Credit Card Fraud Detection (Classification)
- Email/SMS Spam Detection (Classification)
- Regression Problems
### Unsupervised Learning
- Unsupervised learning problems do not have a predictor variable.
- Unsupervised models learn exclusively through unlabelled data. (It could be weak or semi-supervised as well).
- Applications of Unsupervised Learning Algorithms:
- Clustering Algorithms
- Principal Component Analysis
- Recommendation Systems
### Reinforcement Learning
- Teaches AI systems to make decisions through trial and error.
- Applicable in game systems, such as chess or other strategic games.
- For Instance:
We beat the game bot using **x strategic steps** in an online game such as chess. The game could not be beaten with the same **x strategic steps** again, since the game evolves through reinforcement learning.
# Workflow of a ML Problem
_Please note that **accuracy** is not the recommended metric for the evaluation of performance of a machine learning model._
## Moving Forward
| Topic | Concepts | Python Libraries |
| ----------- | ----------- | ----------- |
| [classification](https://github.com/WatashiwaSid/machine-learning-docs/tree/main/classification) | Classification Algorithms | Pandas, Scikit Learn |
| [regression](https://github.com/WatashiwaSid/machine-learning-docs/tree/main/regression) | regression, gradient descent | Scikit Learn, Numpy |
| [clustering](https://github.com/WatashiwaSid/machine-learning-docs/tree/main/clustering) | hard and soft clustering | Scikit Learn, Seaborn |
| [dimensionality reduction](https://github.com/WatashiwaSid/machine-learning-docs/tree/main/dimension-reduction) | pca, low variance, high corelation | Scikit Learn, Seaborn |