I love cats! They are my favorite animals in the entire world. Because of this big love, I have watched a lot of videos about cat's health. And a lot of them said cats are very suspicious about what kind of water we put for them to drink.
The objective of this model is to determine if the water meets the criteria for safe consumption by living beings.
The model was analyzed using pandas. It was imbalanced and had many missing values. So, oversampling was applied, and a pipeline for imputing missing values and standardizing the data was used.
For model training, I selected classification models.
The primary metric was accuracy, with a target score of over 70%.
To compare and record the training, the MLflow API was used.
The models are simple, but Random Forest was able to achieve a prediction accuracy of around 80%.