1. intro-to-pytorch

• these are codes i wrote as i was learning pytorch
• i followed this tutorial: https://machinelearningmastery.com/pytorch-tutorial-develop-deep-learning-models/
• since i didn't have a lot of experience with OOP, some of the code blocks required more time and effort to follow

2. custom_dataset.py

• this code describes how to create a custom dataset in pytorch
• the code was influenced by this tutorial: https://www.journaldev.com/36576/pytorch-dataloader

some key takeaways

• defining a custom dataset requires overwriting __len__ and __getitem__ methods

3. train_dev_split.py

• this is to learn how to split the data into training and dev set using random_split

4. mlp_binaryClassification.py

• code to do binary classification with a multi layer perceptron (mlp)

5. mlp_multiclassClassification.py

• code to do mutli-class classification with a multi layer perceptron (mlp)

iris dataset

• the dataset has 4 inputs and 1 output
• the inputs are:
  • sepal length (cm)
  • sepal width (cm)
  • petal length (cm)
  • petal width (cm)
• the outputs could be either Iris setosa, Iris versicolour, Iris virginica

(image from https://medium.com/@jebaseelanravi96/machine-learning-iris-classification-33aa18a4a983)

6. mlp_regression.py

• code to do regression with a multi layer percept

boston housing

• the dataset has 13 inputs and 1 output
• one of the input is a binary class
• the rest of the inputs are continuous

7. cnn_mnist.py

• code to do handwritten digit recognition (MNIST) with a convolutional neural network (CNN)
• has 2 conv layers and 2 fully connected layers