Becaye BaldéinTowards AIA Trustworthy Model for Loan Eligibility AssessmentA model with high-performance metrics might convince a Data Scientist but is unlikely to earn the trust of domain experts if it can’t…Feb 5Feb 5
Becaye BaldéEfficient Data Preprocessing Sorcery using SklearnLeverage the power of Pipelines and ColumnTransformers to efficiently preprocess data and train models.Aug 9, 2023Aug 9, 2023
Becaye BaldéHow to Use Transformers If You Are a LaymanTransformers can be difficult to train, but luckily for us, we’ve got HuggingFace, a library that provides an easy interface to access…Aug 4, 2023Aug 4, 2023
Becaye BaldéData Tracking Sorcery in Machine Learning using DVCAs data evolves, tracking it becomes harder leading to unreproducible experiments. That’s when DVC comes to the rescue.Jun 19, 2023Jun 19, 2023
Becaye BaldéBayesian Sorcery for Hyperparameter Optimization using OptunaTired of manual tuning, random shots in the dark, or exhaustive grid searches? Bayesian optimization opens the door to a smarter, more…Jun 8, 2023Jun 8, 2023
Becaye BaldéSelf-Organizing Maps ExplainedSelf-organizing maps (SOMs), also known as Kohonen maps, are a type of artificial neural network that are used for clustering…May 15, 20231May 15, 20231
Becaye BaldéWhy is accuracy misleading?While being an intuitive metric, accuracy can be misleading when the data is imbalanced.Apr 14, 2023Apr 14, 2023
Becaye BaldéWhy you should use stratified splitWhen the dataset is imbalanced, a random split might result in a training set that is not representative of the data. That is why we use…Apr 14, 2023Apr 14, 2023
Becaye BaldéVisualize Trends in your Scatter PlotThe LOESS Curve allows us to spot the trends in our scatterplot. In this article, we will create one using R.Apr 1, 2023Apr 1, 2023
Becaye BaldéVisualizing Correlations: Scatter Matrix and Heat map“When performing EDA on a dataset, it is important to visualize correlations. Scatter matrix and heat maps are two of the best ways to…Mar 27, 2023Mar 27, 2023