CST383 Learning Log #6

    This week in class focused on several important machine learning concepts, including linear regression, classification, train/test splits, model evaluation, and hyperparameter tuning. One of the biggest takeaways for me was understanding the difference between training a model and evaluating how well it generalizes to new data. Before this week, I tended to focus on how well a model fit the training data, but I now understand that test performance is a much better indicator of how useful a model will be in practice.

I also learned more about the role of hyperparameters and how they differ from model parameters. The discussion of GridSearchCV helped me understand how machine learning practitioners systematically search for better hyperparameter values rather than choosing them arbitrarily. I found it interesting that the best hyperparameters can vary significantly depending on the dataset being used.


One concept that I’m still working to fully understand is the tradeoff between model complexity and generalization. For example, with KNN classification, I understand that choosing a very small value of k can lead to overfitting, while choosing a very large value can lead to underfitting. However, I sometimes find it difficult to predict exactly how changes in hyperparameters will affect a model before seeing the results. I’d like to develop a stronger intuition for recognizing when a model is too simple or too complex.


Another idea that I found interesting was the use of evaluation metrics such as RMSE, MAE, and accuracy. I kind of understand how to calculate these metrics and what they measure, but I would like to learn more about when one metric is preferred over another in real-world applications.


This week really helped me see machine learning as more than just fitting equations to data. I’m beginning to understand the importance of model evaluation, hyperparameter selection, and testing models on unseen data. These concepts seem essential for building models that are both accurate and reliable.

Comments

Post a Comment

Popular posts from this blog

Computer Science BS Journal (CST334) : Week 3

  This week in CST334 our lectures covered: Address Spaces, C Memory API, Address Translation & Base-and-bounds, Segmentation, Free Space Management. We also started working on our first group project this week. In this learning journal I'll write and explanation/what I think to be the most important bits of each subject. 1. Address Spaces An address space is the range of memory addresses that a process can use. Each process has its own virtual address space, isolated from others, managed by the operating system. Key Takeaways: Provides process isolation . Helps with security and stability . Enables virtual memory and memory protection. 2. C Memory API The C memory API includes functions for dynamic memory allocation and deallocation: malloc() , calloc() , realloc() , and free() . Key Takeaways: malloc(size) allocates a block of memory. calloc(n, size) allocates and zero-initializes memory. realloc(ptr, new_size) resizes previously allocated memory. ...
Read more

Computer Science BS Journal (CST334) : Week 5

    This week in CST334 our lectures covered: Concurrency and Threads, Threads API, Locks, and Lock-based data structure. In this learning journal I'll write and explanation/what I think to be the most important bits of each subject. We also had our midterm this week and I didn't do as good as I wanted to, I definitely need to focus some more on my studies. Concurrency and Threads Concurrency is the ability of a system to handle multiple tasks at the same time. Threads are the smallest units of execution within a process. Concurrency allows threads to run independently, improving efficiency, especially on multi-core processors. Key Takeaways: Concurrency enables better CPU utilization. Threads share the same memory space within a process. Context switching allows threads to appear to run in parallel. Can lead to race conditions if not properly synchronized. Threads API Threads APIs provide the functions and mechanisms to create, manage, and synchronize threads....
Read more

Computer Science BS Journal (CST334) : Week 2

    This week in CST334 our lectures covered: Processes, C Process API, Limited Direct Execution, CPU Scheduling, and MLFQ. In this learning journal I'll write and explanation/what I think to be the most important bits of each subject. Processes A process is a running instance of a program. It includes: Code (instructions) Data (variables, heap) Execution context (registers, stack, program counter) Operating systems manage processes to provide isolation, resource control, and scheduling. C Process API The C Process API refers to system calls and library functions in C for process management. Key functions include: fork() – Creates a new process by duplicating the current one. exec() – Replaces the current process image with a new one. wait() – Waits for a child process to finish. exit() – Terminates the current process. Limited Direct Execution (LDE) Limited Direct Execution is a technique where the OS allows a user program to run directly ...
Read more