Suhaib's CS Journal

Learning Journal Week 1: Before this course, I expected Software Engineering to mostly be about writing code and learning how to build bigger programs. I thought the class would focus mainly on programming, using tools, and maybe working on projects like a real software developer. After completing week 1, my opinion has changed because I realized Software Engineering is more than just coding. It also includes planning, testing, debugging, security, APIs, and making sure software is reliable and maintainable. The labs showed me that writing the code is only one part of the process. Testing the code and proving that it works is just as important. I also learned that software engineers need to think about how different parts of an application connect, like controllers, repositories, databases, and security. I now see Software Engineering as a process for building software correctly, not just getting something to run.

Learning log week 7: This week we covered encoding categorical variables, logistic regression, and overfitting. I learned that categorical variables need to be changed into numerical values before most machine learning models can use them. One hot encoding made sense to me because it creates separate columns for each category instead of assigning numbers that could accidentally suggest an order. Logistic regression was a little confusing at first because the name sounds like it should predict a numerical value, but it is actually used for classification. I understand now that it predicts the probability of an outcome, such as whether a customer will churn or not. I still want more practice interpreting the coefficients and understanding exactly how they affect the predicted probability. The topic of overfitting also stood out to me. A model can perform very well on the training data but still perform poorly on new data. This helped me understand why we need test sets and cross valida...

Learning Log | Week 6 This week we learned about hyperparameter tuning, KNN regression, linear regression, and how to evaluate regression models. Hyperparameter tuning helped me understand that some model settings are chosen before training and that the best values depend on the dataset. Grid search seems useful because it tests different combinations in an organized way, but I still wonder how to choose a reasonable range of values without testing too many combinations. The difference between KNN regression and linear regression was also interesting. KNN uses nearby training examples to make predictions, while linear regression learns coefficients that describe the relationship between the predictors and the target. Linear regression seems faster when making predictions, but it may struggle when the relationship is not close to linear. The hardest part for me was interpreting regression errors and predicted versus actual plots. I understand that lower MSE and MAE values are better, ...

Learning log 4: This week we learned more about how important preprocessing is before building a machine learning model. One thing that stood out to me is that missing data is not always obvious. Sometimes it shows up as actual NA values, but other times it can be hidden as values like 0 or “information requested,” depending on the dataset. That made me realize that cleaning data is not just a technical step, but also requires thinking carefully about what the values actually mean. I also learned why scaling matters, especially for models like KNN. Since KNN uses distance to compare points, features with larger numbers can have too much influence if the data is not scaled. This helped me understand why preprocessing and modeling are connected instead of being separate tasks. The test/train split and cross validation topics were useful too. I understand that the test set should be saved for checking how well the model works on new data, while cross validation helps compare models or hyp...