Basic_Machine_learning

Simple Regression

This section focuses on implementing a simple regression model using quadratic equations and optimizing it with PyTorch.

• Function Creation: Define the target function as target = Ax² + Bx + C and the prediction function as predict = A'x² + B'x + C'.
• Loss Function: Implement the Mean Squared Error (MSE) Loss as MSELoss = 0.5 * (target - predict)².
• Regression with PyTorch: Utilize the PyTorch API to minimize the loss and make the prediction as close as possible to the target.

LeNet-5 with MNIST

In this section, we implement and train the LeNet-5 neural network using the MNIST dataset.

• Neural Network Creation: Construct the LeNet-5 architecture using PyTorch API.
• Training: Train the network using the MNIST dataset.
• Testing: Evaluate the training results and model performance.

Custom Dataset

Working with a custom dataset, training models like VGG-16 or ResNet18.

• Dataset Access: Load and preprocess the custom dataset using PyTorch API.
• Model Training: Train models like VGG-16 or ResNet18 on the custom dataset.

---

Each section provides practical insights into the basics of machine learning and deep learning, leveraging PyTorch framework.