Q 12) Difference between correlation and covariance? When should we prefer one over the
other?
Correlation is a function of the covariance. Covariance is used
to determine how much two random variables vary together, whereas
correlation is used to determine when a change in one variable can
result in a change in another. Covariance gives the extent to which
two random variables change. Correlation represents how strongly two
random variables are related.
Tho not a rule, but is preferred to use the covariance matrix when the
variable are on similar scales and the correlation matrix when the
scales of the variables differ.
13.) When should we do chi sq test? what is it used for?
A Chi-Square test is a test of statistical significance for
categorical variables. We typically use it to find how the observed
value of a given event is significantly different from the expected
value. Intuitively, we see whether our categorical feature is
relevant/significant to predict the output variable. we use it for
feature selection. We renove values that are not so significant ace to
test.
Q 14, If both fiter kemels and pooling layers extract features from an image in CNN, then
what is the difference between them?
Filters extract information from the image like Edges, Patters,
etc. This info that needs to be extracted are learned during training.
Tt depends largely on Data what info needs to be extracted fron
filters. Whereas Pooling does not extract features, rather it takes the
avg value of nearby n pixels so that the most important info in the
image is retained reducing the image size. which is req to make large
models.
Q.15,) Difference between vectors and scalers?
Scalars are quantities that are fully described by a magnitude
(or numerical value) alone. Vectors are quantities that are fully
described by both a magnitude and a direction.