6/ Logistic regression is highly interpretable. Each one-hot encoded category gets its own weight: a positive weight pushes the predicted churn probability up, while a negative weight pushes it down. #mlzoomcamp

5/ Machine learning models only work with numbers, making categorical string values useless on their own. One-hot encoding solves this by replacing a categorical column with several binary columns, turning each category into a 0 or 1. #mlzoomcamp

4/ For numerical features, the Pearson correlation coefficient measures the degree of dependency with the target. A negative correlation, like with customer tenure, means the longer a customer stays, the less likely they are to churn. #mlzoomcamp

3/ Measuring the importance of categorical variables requires mutual information. This concept from information theory tells us exactly how much we learn about the target churn variable by knowing a specific feature's value. #mlzoomcamp

2/ Logistic regression builds directly on linear regression. By applying a sigmoid function to the weighted sum of features, it squashes any real number score into a probability between 0 and 1. #mlzoomcamp

Kicking off Module 3 of #mlzoomcamp. We are shifting from regression to classification by predicting telecom customer churn. The goal is to compute a probability score for each customer to identify who is likely to leave before they actually do. #mlzoomcamp

Another regression homework completed! 🚀 Step by step, pandas, NumPy, linear regression, RMSE, regularization, and data preparation are starting to make much more sense.
#mlzoomcamp #LearningInPublic #DataScience

After choosing the final model, I learned that we can combine the training and validation datasets, train again on more data, and then evaluate once on the test set.
#mlzoomcamp #MachineLearning

I learned that a random seed makes a shuffled dataset split reproducible. Using the same seed means we can reproduce the same train/validation/test split later. 🔁
#mlzoomcamp #MachineLearning

Today I compared validation RMSE for different regularization values. Testing different values helped me understand that model settings should be evaluated with validation data, not test data.
#mlzoomcamp #DataScience

A small but useful lesson from #mlzoomcamp: always check your dataset for missing values before training a model. In pandas, df.isnull().sum() makes this really easy.
#MachineLearning #Python

Practiced working with NumPy arrays for linear regression today. I’m getting more comfortable with matrix operations like transpose, dot products, and matrix inversion. 📊
#mlzoomcamp #Python #MachineLearning

Today I learned why we need separate training, validation, and test datasets. Training is for learning, validation helps us choose the model, and test data is for the final evaluation.
#mlzoomcamp #MachineLearning

Finished my Regression homework for Machine Learning Zoomcamp! 🎉 I practiced data preparation, linear regression, RMSE, missing values, random seeds, and regularization.
#mlzoomcamp #MachineLearning

Today I practiced regularized linear regression and tested different values of r. It was interesting to see how regularization changes the validation RMSE.
#mlzoomcamp #MachineLearning

I experimented with different random seeds today and saw that changing the seed changes the train/validation split and can slightly change the model's RMSE.
#mlzoomcamp #MachineLearning

An important lesson from my regression homework: when filling missing values with the mean, calculate the mean only from the training data. Using validation or test data can cause data leakage.
#mlzoomcamp #DataScience

Today I learned more about RMSE (Root Mean Squared Error). It helps us measure how far our model's predictions are from the actual values. Lower RMSE means better predictions.
#mlzoomcamp #MachineLearning

One thing I learned today: missing values can affect a machine learning model. I practiced filling missing values with 0 and with the mean, then compared the RMSE results.
#mlzoomcamp #MachineLearning

Today I practiced Linear Regression in the Machine Learning Zoomcamp. I learned how to split a dataset into training, validation, and test sets using NumPy.
#mlzoomcamp #MachineLearning