https://github.com/guillainbisimwa/udacityml
Intro to Machine Learning with Sebastian and Katie
https://github.com/guillainbisimwa/udacityml
Last synced: 7 months ago
JSON representation
Intro to Machine Learning with Sebastian and Katie
- Host: GitHub
- URL: https://github.com/guillainbisimwa/udacityml
- Owner: guillainbisimwa
- Created: 2019-10-16T16:04:06.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2022-06-21T23:08:41.000Z (over 3 years ago)
- Last Synced: 2025-01-30T05:42:45.591Z (9 months ago)
- Language: DIGITAL Command Language
- Size: 18.1 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
ud120-projects
==============
# Udacity's Machine Learning Nanodegree project files and lecture notes.
This repository contains project files for the Introduction to Udacity's Machine Learning Engineer Nanodegree program.
## Model evaluation and validation
Topics covered in this section:
### Model Evaluation
Confusion matrix, F1 score, F-beta score, ROC curve
### Model selection
Types of errors, various types of cross validation, learning curves, grid search
## Supervised learning
Topics covered in this section:
- Linear regression
Absolute trick, advantages / disadvantages, L1 regularisation, L2 regularisation
- Decision trees
Entropy, information gain, hyperparameters
- Naive bayes
Prior probability, posterior probability, naive bayes
- Support vector machines (SVM)
Idea, different types of errors, basic working principle, etc.
## Unsupervised learning
Topics covered in this section:
- Clustering
K-means clustering
- Hierarchical and density-based clustering
Hierarchical clustering, single-link clustering, complete-link clustering, average-link clustering, ward's method, DB scan
- Gaussian mixture model and cluster validation
EM algorithm, cluster validation, external indices, internal indices, adjusted rand indices, silhouette coefficient
- Feature scaling
- PCA
- Random projection and ICA
## Deep learning
Topics covered in this section:
- Neuronal networks
Perceptron trick, perceptron algorithm, sigmoid activation, maximum likelihood, cross entropy, logistic regression, perceptron and gradient descent
- Deep neural networks
Regularization, dropout, vanishing gradients and activation function, momentum, keras optimisers
- Convolutional neural networks
Model validation, image augmentation
## Reinforcement learning
Topics covered in this section:
- RL framework
Reinforcement setting, episodic and continuous tasks, rewards hypothesis, cumulative reward, discounted reward, Markov decision process, Bellman equations, optimality, action-value functions,
- Dynamic programming
Iterative policy evaluation, estimation of action values, policy improvement, policy iteration, truncated policy iteration, value iteration
- Monte Carlo methods
Predicting state values, estimating action-values, incremental mean, policy evaluation, policy improvement, exploration-exploitation dilemma, GLIE MC control algorithm, constant-alpha GLIE MC control algorithm
- Temporal difference learning
TD(0) prediction, action value estimation, solving the control problem, Sarsamax (Q-learning), expected Sarsa
- Deep reinforcement learning
Discrete and continuous spaces, discretization, coarse coding, tile coding, function approximation, kernel functions, coarse coding
- Deep Q-Learning
NNs as value functions, Monte Carlo learning, TD learning, Q-learning, Sarsa vs. Q-learning, experience replay, fixed Q-targets, different types of DQNs
- Policy-based methods
Policy function approximation, stochastic policy search, policy gradients, Monte Carlo policy gradients, constrained policy gradients
- Actor-critic methods
## Author
👤 **Guillain Bisimwa**
- Github : [@guillainbisimwa](https://github.com/guillainbisimwa)
- Twitter : [@gullain_bisimwa](https://twitter.com/gullain_bisimwa)
- Linkedin : [guillain-bisimwa](https://www.linkedin.com/in/guillain-bisimwa-8a8b7a7b/)
## 🤝 Contributing
Contributions, issues, and feature requests are welcome!
## Acknowledgments
- UDACITY
## Show your support
Give a ⭐️ if you like this project!