PinnedAli S·Feb 8, 2023Grid search and random search are outdated. This approach outperforms both.If you’re a data scientist, there is a good chance you have used “Grid Search” to fine-tune the hyperparameters of your model. This is a…A response icon6A response icon6
PinnedInTDS ArchivebyAli S·Apr 2, 2022Stop Using SMOTE to Treat Class Imbalance. Take This Intuitive Approach InsteadLet’s expose the myth; SMOTE doesn’t deserve the hype. This simple approach does!A response icon13A response icon13
PinnedAli S·Feb 25, 2022Stop using random forest feature importances. Take this intuitive approach insteadA more accurate and intuitive alternative to random forest feature importancesA response icon19A response icon19
InTDS ArchivebyAli S·Apr 8, 2022Your validation loss is lower than your training loss? This is why!Just because your model learns from the training set doesn’t mean its performance will be better on it.A response icon3A response icon3
Ali S·Mar 24, 2022Basics of Network Theory that every data scientist must know in simple termsData science is more than just machine learning. You need to know this much about networks too!A response icon2A response icon2
InTDS ArchivebyAli S·Mar 17, 20225 Facts they don’t tell you about random forest feature importances5 Key facts to know before interpreting tree-based feature importancesA response icon1A response icon1
Ali S·Mar 11, 2022What is K-Medoids Clustering and When should you use it instead of K-MeansThere are many unsupervised learning algorithms out there, from the famous K-Means and hierarchical methods to Gaussian Mixture models and…
