DheemanthBhat / ML-Concepts

Notebooks explaining various Machine Learning concepts.

ML-Concepts

Index

1. Distributions.
2. EDA.
3. KNN Algorithm.
4. Naive Bayes classifier.
5. Logistic Regression.
6. Linear Regression.
7. Support Vector Machine.
8. Decision Tree
9. Random Forest
10. Boosting

1 Distributions

1. Normal (Gaussian) Distribution
2. Central Limit Theorem
3. Q-Q PLot

2 EDA: Exploratory Data Analysis

1. EDA of Haberman's Survival dataset
2. EDA of Titanic Mortality dataset.

3 K-Nearest Neighbor (KNN)

Note: Create folder by name Output inside 3. K-Nearest Neighbor (KNN) folder before funning the notebook.

Objective of K-Nearest Neighbor (KNN).ipynb notebook:

1. Plot cluster of data-point.
2. Implementing KNN algorithm from scratch.
3. Test the accuracy of the model trained on KNN.

4 Naive Bayes classifier

Objective of Naive Bayes Classifier.ipynb notebook:

1. Implement simple Naive Bayes classifier for categorical features.
2. Train and test model on sample weather forecast dataset using Naive Bayes.

Objective of Naive Bayes Classifier - CategoricalNB.ipynb notebook:

1. Train Naive Bayes classifier using CategoricalNB from sklearn.
2. Train and test model on sample weather forecast dataset.

5 Logistic Regression

Objective of 3. Logistic Regression.ipynb notebook:

Implement multinomial Logistic Regression using Gradient Descent minimization technique.

6 Linear Regression

Objective of 1. Squared Error Cost Function.ipynb notebook:

Implement cost function for multivariate Linear Regression based on Squared Error technique.

Objective of 2. Batch Gradient Descent.ipynb notebook:

Implement Batch Gradient Descent to minimize Squared Error Cost function.

Objective of 3. Linear Regression - Gradient Descent.ipynb notebook:

Implement multivariate Linear Regression using Gradient Descent minimization technique.

Objective of 4. Linear Regression - Normal Equation.ipynb notebook:

Implement multivariate Linear Regression using Normal Equation technique.

7 Support Vector Machine (SVM)

Objective of 1. Linear SVR - Gradient Descent.ipynb notebook:

Implement Support Vector Regression without any Kernel i.e., linear-SVR.

8 Decision Tree

Objective of 2. Decision Tree - DecisionTreeClassifier.ipynb notebook

Visualize decision tree