Machine Learning for Everyone (Mark Fenner)

18 flashcards · Computer Science · by lakesare

1. Attribute VS value VS feature? A simple biomedical data table. Each row is an example. Each column contains values for a given attribute. Together, each attribute-value pair is a feature of an example.

   attribute - column name (height) what-is-measuredvalue - column value (5'2'') what-the-value-isfeature - column name and value ({ height: 5'2'' }) what-the-measured-value-is

2. When is the process of learning the relationship between the input and the target feature called classification VS regression? Feature is a column name and value ({ height: 5'2'' }). In any case, we can pick a concrete, measurable target and ask, “Can we find a predictive relationship between the attributes we have today and the outcome that we will see at some future time?” We are literally trying to predict the future—maybe ten years from now—from things we know today. We call the concrete ou…

   classification - target feature is discreet. We call the concrete outcome our target feature or simply our target. If our target is a category like { sick, healthy } or { None, I, II, III }, we call the process of learning the relationship classification. Here, we are using the term classification in the sense of finding the different classes, or categories, of a possible outcome.regression - target feature is a smooth sweeping of numerical values. If the target is a smooth sweeping of numerica…

3. Where does the "regression" term come from?

   From Galton's "regression towards the mean". The term "regression" was used by Francis Galton in his 1886 paper "Regression towards mediocrity in hereditary stature". Francis Galton (UK, 1822 - 1911) - a polymath who made important contributions in many fields of science.

4. What did Galton call regression towards the mean? Francis Galton (UK, 1822 - 1911)

   Children of tall parents regress to the average height of an adult. Based on these results, Galton concluded that as heights of the parents deviated from the average height (that is as they became taller or shorter than the average adult), their children tended to be less extreme in height.

5. What's linear regression?

   When we have the dots, and we want to find a line that fits these dots.

6. How is linear regression connected to "regression" towards the mean?

   Galton used linear regression analysis when he was describing regression towards the mean. The term was coined by Sir Francis Galton when he noticed that tall parents tend to have children shorter than them, whereas short parents often have children who were taller than them. Graphically, this means that if we plot the height of parents on the x-axis and the height of kids on the y-axis and draw a line through all the data points, we get a line with a slope of less than 1 (or equivalently an an…

7. [Classifiers VS regressors] Name 1 example of a classifier neural network. Classifiers are models that take input examples and produce an output that is one of a small number of possible groups or classes

   Image classification: from an input image, output the animal (cat, dog, zebra, none) that is in the image.

8. [Classifiers VS regressors] Name 1 example of a regressor neural network. Numerical values surround us in modern life. Physical measurements (temperature, distance, mass), monetary values, percents, and scores are measured, recorded, and processed endlessly. Each can easily become a target feature that answers a question of interest.

   Student score: from class attendance, etc. output the student score. As with many regression problems, we could reasonably convert this regression problem to a classification problem by predicting a pass/fail or a letter grade instead of a raw numerical score.

9. What does sklearn stand for?

   Scikit-learn. A side note: scikit-learn is the project’s name; sklearn is the name of the Python package.People use them interchangeably in conversation.

10. Probability distribution is a mapping between _________ _________?

    Probability distribution is a mapping between events and their probabilities. If you give me an event, then I can look it up in the probability distribution and tell you the probability that it occurred.

11. Name 2 probability distributions.

    Uniform distribution When a group of events shares a common probability value—like the different faces on a fair die—we call it a uniform distribution.Normal distribution 1. Its midpoint has the most likely value—the hump in the middle.2. It is symmetric—can be mirrored—about its midpoint.3. As we get further from the midpoint, the values fall off more and more quickly

12. [Numpy] What will np.arange() do?

    It's like python's range() which returns a list, but returns an array.

13. What does this denote?

    Expected value of a game p - probabilities of the events andv - being the values or payoffs of those events

14. [Python] How to deal with this in python?

    Here, numpy is checking if the numbers are the same for many, many decimal places—out to the point where the difference is insignificant. If we care, we can define our own tolerance for what is and isn’t significant.

15. In the normal distribution, what are other names for: center spread?

    center - mean (μ) spread - standard deviation (σ)

16. Sepal VS petal in an iris?

17. What does k-NN stand for?

    k-Nearest Neighbors

18. Name 1 example of nonparametric learning method.

    k-NN (k-Nearest Neighbors)