Pytorch Deep Learning Outline
Programming for Data Science is a subject we’ve designed to explore the various programming components of data science.
Keywords
data science, data analysis, programming, dataidea
Week 1: Tensors and Datasets
- 1 Dimension Tensors
- Two Dimension Tensors
- Derivatives and Graphs in PyTorch
- Simple Dataset
- Pre Built Datasets
Week 2: Linear Regression
- Linear Regression 1 Dimension
- Linear Regression with 1 Parameter
- Training Slope and Bias
Week 3: Linear Regression in PyTorch
- Stochastic Gradient Descent
- Mini-Batch Gradient Descent
- PyTorch Build-in Functions
- Training and Validation Sets
Week 4: Multiple Input Linear Regression
- Making Predictions in Multiple Linear Regression
- Training a Multiple Linear Regression Models
- Multi-Target Linear Regression
- Training Multiple Output Linear Regression Models
Week 5: Logistic Regression
- Making Predictions in Multiple Linear Regression
- Logistic Regression and Bad Initialization Values
- Cross Entropy Loss Function
- Sofmax Activation in 1 Dimension
Week 6: Practice
- Practice
Week 7: Shallow Neural Networks
- Simple One Hidden Layer
- Multiple Neurons
- Noisy XO
- One Layer Neural Network
- Activation Functions
- Test Activation Functions
Week 8: Deep Neural Networks
- Multiple Linear Regression
- Deeper Neural Networks with nn.ModuleList()
- Using Dropout for Classification
- Neural Networks with Momentum
Week 9: Convolution Neural Networks
- What is Convolution
- Activatioin Function and Max Pooling
- Multiple Channel Convolutional Neural Network
- Convolutional Neural Network with Batch Normalization Get Started
Week X: Capstone Project and Review
- Applying learned concepts to a real-world dataset