Tre rilevatori di novità

Finalmente sai abbastanza per eseguire alcuni test sull’uso di alcuni rilevatori di anomalie sul dataset di aritmia. Per valutarne le prestazioni, li allenerai su un insieme di dati di training senza etichette e poi confronterai le loro previsioni con la verità a terra nei dati di test usando il metodo .score_samples(). Questa volta ti verrà chiesto di importare i rilevatori come parte dell’esercizio, ma i dati X_train, X_test, y_train, y_test sono già precaricati come al solito.

Questo esercizio fa parte del corso

Progettare workflow di Machine Learning in Python

Visualizza il corso

Esercizio pratico interattivo

Prova a risolvere questo esercizio completando il codice di esempio.

# Import the novelty detector
from sklearn.____ import ____ as onesvm

# Fit it to the training data and score the test data
svm_detector = ____.____(X_train)
scores = svm_detector.____(X_test)

Modifica ed esegui il codice

Questo esercizio fa parte del corso

Progettare workflow di Machine Learning in Python

AvançadoNível de habilidade

4.8+

84 reviews

Inizia il corso gratis

In this chapter, you will be reminded of the basics of a supervised learning workflow, complete with model fitting, tuning and selection, feature engineering and selection, and data splitting techniques. You will understand how these steps in a workflow depend on each other, and recognize how they can all contribute to, or fight against overfitting: the data scientist's worst enemy. By the end of the chapter, you will already be fluent in supervised learning, and ready to take the dive towards more advanced material in later chapters.

Exercise 1: Supervised learning pipelines Exercise 2: Feature engineering Exercise 3: Your first pipeline Exercise 4: Model complexity and overfitting Exercise 5: Grid search CV for model complexity Exercise 6: Number of trees and estimators Exercise 7: Feature engineering and overfitting Exercise 8: Categorical encodings Exercise 9: Feature transformations Exercise 10: Bringing it all together

In the previous chapter, you perfected your knowledge of the standard supervised learning workflows. In this chapter, you will critically examine the ways in which expert knowledge is incorporated in supervised learning. This is done through the identification of the appropriate unit of analysis which might require feature engineering across multiple data sources, through the sometimes imperfect process of labeling examples, and through the specification of a loss function that captures the true business value of errors made by your machine learning model.

Exercise 1: Data fusion Exercise 2: Is the source or the destination bad?Exercise 3: Feature engineering on grouped data Exercise 4: Imperfect labels Exercise 5: Turning a heuristic into a classifier Exercise 6: Combining heuristics Exercise 7: Dealing with label noise Exercise 8: Loss functions Part I Exercise 9: Reminder of performance metrics Exercise 10: Real-world cost analysis Exercise 11: Confusion matrix calculations Exercise 12: Loss functions Part II Exercise 13: Default thresholding Exercise 14: Optimizing the threshold Exercise 15: Bringing it all together

In the previous chapter, you employed different ways of incorporating feedback from experts in your workflow, and evaluating it in ways that are aligned with business value. Now it is time for you to practice the skills needed to productize your model and ensure it continues to perform well thereafter by iteratively improving it. You will also learn to diagnose dataset shift and mitigate the effect that a changing environment can have on your model's accuracy.

Exercise 1: From workflows to pipelines Exercise 2: Your first pipeline - again!Exercise 3: Custom scorers in pipelines Exercise 4: Model deployment Exercise 5: Pickles Exercise 6: Custom function transformers in pipelines Exercise 7: Iterating without overfitting Exercise 8: Challenge the champion Exercise 9: Cross-validation statistics Exercise 10: Dataset shift Exercise 11: Tuning the window size Exercise 12: Bringing it all together

In the previous chapters you established a solid foundation in supervised learning, complete with knowledge of deploying models in production but always assumed you a labeled dataset would be available for your analysis. In this chapter, you take on the challenge of modeling data without any, or with very few, labels. This takes you into a journey into anomaly detection, a kind of unsupervised modeling, as well as distance-based learning, where beliefs about what constitutes similarity between two examples can be used in place of labels to help you achieve levels of accuracy comparable to a supervised workflow. Upon completing this chapter, you will clearly stand out from the crowd of data scientists in confidently knowing what tools to use to modify your workflow in order to overcome common real-world challenges.

Exercise 1: Rilevamento di anomalie Exercise 2: Un semplice outlier Exercise 3: Contaminazione in LoF Exercise 4: Rilevamento di novità Exercise 5: Una semplice novelty Exercise 6: Tre rilevatori di novità

Esercizio in corso

Exercise 7: Contamination, di nuovo Exercise 8: Apprendimento basato sulla distanza Exercise 9: Trova il vicino Exercise 10: Non tutte le metriche sono d'accordo Exercise 11: Dati non strutturati Exercise 12: Levenshtein ristretto Exercise 13: Mettere tutto insieme Exercise 14: Considerazioni finali