Crowdly
Додати до Chrome
2026_ML pour la reconnaisance des formes
Шукаєте відповіді та рішення тестів для 2026_ML pour la reconnaisance des formes? Перегляньте нашу велику колекцію перевірених відповідей для 2026_ML pour la reconnaisance des formes в moodle.epita.fr.
Отримайте миттєвий доступ до точних відповідей та детальних пояснень для питань вашого курсу. Наша платформа, створена спільнотою, допомагає студентам досягати успіху!
- given the following descriptors extracted from this image: , , , , , , , , , , , ,
- and given the result of the clustering in the descriptor space as illustrated in the image above (letter: cluster name and order, stars: cluster centers, lines: clusters boundaries).
- order vector values in the proper cluster order
- write only values, separated by a space.
For instance, to write the vector , write "1 5 -4 0".
The figure above depicts a distribution of "false" and "true" classes given a certain feature.
The figure also shows several potential places (a, b, c, d, e, f, g) where a threshold could be set.If the "true" class indicates a poisonous mushroom, which threshold should you use to be safe?Suppose you have an imbalanced dataset where we have the following classes and prior probabilities:
- A: 0.6
- B: 0.1
- C: 0.25
- D: 0.05
What is the value of the best accuracy that a dummy classifier would get?
The figure above depicts a distribution of "false" and "true" classes given a certain feature.
The figure also shows several potential places (a, b, c, d, e, f, g) where a threshold could be set.Select among the following curves the one which represents the right "Receiver Operating Characteristic" (ROC) curve corresponding to this distribution and thresholds.
What is the expected accuracy of a truly random classifier (which picks one of the possible classes with a uniform probability) if we have 6 classes with equal probabilities.
Please provide 2 significant digits.
Match each classifier category with its definition.
- given the following descriptors extracted from this image: , , , , , , , , , , , ,
- and given the result of the clustering in the descriptor space as illustrated in the image above (letter: cluster name and order, stars: cluster centers, lines: clusters boundaries).
- order vector values in the proper cluster order
- write only values, separated by a space.
For instance, to write the vector , write "1 5 -4 0".
When training an image classifier, what do we need?
Match each pipeline with its definition.