Machine Learning and Data Analysis Course Description

۱) Data and EDA

a.  types of data

b. EDA Best Practices

 

۲) Preprocessing

a. Missing Values and Imputation → Exercise Se۰۲_۰۱ , Exercise Se۰۲_۰۲

b. Scaling Data → Exercise Se۰۲_۰۳

c. Categorical Encoding

 

  Project ۱

---

۳) Probability

a. Some Key Concepts, Descriptive statistics, Distribution, Central tendency, Variability

b. Inferential statistics, Descriptive statistics

c. Basic Probability

• i. Axiomatic Probability → Exercise cho۱_۱

• ¡i. Conditional Probability, Bayes' Theorem or Rule → Exercise cho۱_۲

d. Random Variables and Distribution

• i. Random variable

• ii. Cumulative distribution function (cdf)

• iii. Probability mass function (pmf)

---

e. Distribution

f. Goodness of fit and Hypothesis test (dist_fitter) → Exercise Ses_۰۴_۱

g. Expectation of a Random Variable → Exercise Ses_۰۴_۲

h. Variance and Standard Deviation

i. Quantile

j. Skewness and Kurtosis

 

۴) ML Fundamentals

a. Defining

b. Generalized Model

• i. Overfitting

• ii. Underfitting

• iii. Bias, Variance

 

---

۵) Validation Strategy

a. Hold-Out

b. K-fold

c. Startified

 

۶) Metrics

a. Accuracy

b. Precision

c. Recall

d. ROC_AUC score

e. R' (R-Squared)

f. Mean Squared Error (MSE)

g. Root Mean Squared Error (RMSE)

h. Mean Absolute Error (MAE)

   Practice

 

---

۷) Linear Regression

a. The equation of a line

b. Simple Linear Regression

c. Measures of Variation

d. A simple Example → Jupyter Notebook → Exercise S۰۵_E۰۱

 

۸) Polynomial Regression

a. Nonlinear functions

b. Higher-order polynomials

c. Overfitting and complexity

    Practice

 

---

۹) Logistic Regression

a. The Logistic Function

b. Dilemma using OLS

c. Odds ratio

d. Probability Thresholds

e. Cost Function

f. Optimization Process → Exercise S۰۶_۰۱

 

۱۰) Regularization

a. Overfitting, Underfitting

b. Occam's razor

c. Bias-Variance Tradeoff

d. Model Selection

e. Regularization (Explain)

f. Ridge & Lasso Regression → Excersice S۰۶_۰۲

 

۱۱) Classification Metrics

a. accuracy_score

b. confusion_matrix

c. recall_score

d. precision_score

e. fi_score

f. roc_curve

g. roc_auc_score

h. precision_recall_curve

i. average_precision_score(PR AUC)

j. log loss

  Project ۲:

  Full Pipeline Practice (Covid-۱۹ Hospital los)   

 

---

۱۲) Tree Based Algorithms

a. Decision Trees (CART)

b. Random Forest

c. AdaBoost

d. GBM

e. XGBoost

 

۱۳) Neural Networks

a. Biological Inspirations

b. Introduction

c. Elements

d. Matrix Operation

e. Activation Functions

f. Training

g. Gradient Descent Algorithm

h. Adam

i. Loss Functions

j. Generalization

k. Batch Normalization

l. Keras → Exercise_ MNIST

m. Convolutional Neural Networks

n. Recurrent Neural Networks

o. LSTM Networks

---

۱۴) Unsupervised Learning

a. Cluster Analysis

b. DBSCAN

c. Esp and Min Point

d. K-Means

e. PCA

---

۱۵) Reinforcement Learning (Optional)