How did you split the sample into training and testing?
Machine Learning / AI
What are Medusa and multi-token prediction heads?
What are Sentence Embeddings and how do they differ from word embeddings?
What is Bayes' theorem?
What methods are used for selecting negative examples?
How is the probability distribution of a normal distribution arranged in binary classification?
What popular AutoML solutions are there (H2O, AutoGluon, FLAML, AutoKeras)?
What is Data Drift?
What practices are used when backfilling historical features?
How do you behave in a team, and how conflict-prone are you?
If you multiply all model scores by 2, will the ROC AUC change?
How does the amount of data affect the choice of search index?
What are the pitfalls when comparing multimodal models on benchmarks?
What was your profession?
What is targeting?
Who do you most often communicate with at work?
What is Dreambooth and how to avoid overfitting?
How to split data into strata for validation?
What is the difference between sentiment classification at the sentence level and at the aspect level?