Machine learning basics and applications, especially about the supervised learning.

1. Machine_Learning_HW1
2. Machine_Learning_HW2
3. Machine_Learning_HW3

In this project, we will discuss the basic concepts that are used to implement the linear regression model: gradient descent and least square. Then, we will discuss how we evaluate the performance of the prediction model by evaluating the error: mean squared error (MSE). After discussing all the concepts, we will present the implemented python code for task 1 and task 2, which represents each model that applies gradient descent and the least squaremethod of the linear regression model. In the result, we will show the prediction graph according to the given data set and the coefficients for the predicted model. Also, there will be comparisons between the built-in linear regression models and the implemented models.

• Visualization of the linear regression model of task 1 with the original data set

• Visualization of the linear regression model of task 2 with the original data set

• Building Environment:

1. The environment that can open the ipython notebook: COLAB, Jupyter Notebook
2. Each cell in the notebook should be executed sequentially.

• File Upload: Before cell execution, the given data set must be uploaded first.

In this project, we will discuss the basic concepts that are used to implement the logistic regression model: sigmoid, logistic regression, hyperbolic function, decision boundary, and underfitting and overfitting. Then, we will discuss how we evaluate the performance of the prediction model by evaluating the error: cross entropy function. After discussing all the concepts, we will present the implemented python code for task 1 and task 2, which represents training the model with the training data set and classifying the data in the test data set using the trained model. In the result, we will show the prediction graph according to the given data set and the coefficients for the predicted model. Also, there will be comparisons between the built-in logistic regression models and the implemented models.

• Classification for the train data set by implemented logistic regression model

• Classification for the test data set by built-in polynomial logistic regression model

Building Environment:

1. The environment that can open the ipython notebook: COLAB, Jupyter Notebook
2. Each cell in the notebook should be executed sequentially

• File Upload: Before cell execution, the given data set must be uploaded first.

In this project, we will discuss the basic concepts that are used to implement the neural network model: perceptron model, activation functions, multi-layer perceptron (MLP), MLP regressor, chain rule, and backpropagation. Then, we will discuss how we evaluate the performance of the prediction model by evaluating the error: mean squared error (MSE). After discussing all the concepts, we will present the implemented python code of multiple classes that are used in the MLP regressor, which represents training the model with the training data. In the result, we will show the prediction graph according to the given data set and the coefficients for the predicted model. Also, there will be comparisons between the built-in logistic regression models and the implemented models

• Result of the multi-perceptron (MLP) model training and its visualization:

• Weights of each layer in the multi-layer perceptron (MLP) model:

• Building Environment:

1. The environment that can open the ipython notebook: COLAB, Jupyter Notebook
2. Each cell in the notebook should be executed sequentially

• File Upload: Before cell execution, the given data set must be uploaded first.