Uda

Question 1

Let’s … a bit.

Answer 1

sidetrack

Question 2

he does not let himself get … by trends.

Answer 2

sidetracked

Question 3

LDA assumes that the every chunk of text we feed into it will contain words that are somehow related. Therefore choosing the right corpus of data is …

Answer 3

crucial

Question 4

Remove very rare and very … words.

Answer 4

common

Question 5

Therefore, common words like “the” and “for”, which appear in many documents, will be .. down. Words that appear frequently in a single document will be scaled up.

Answer 5

scaled

Question 6

A … but effective device from the 1840’s, rather like a thermometer. It works, but a human must estimate its readings.

Answer 6

primitive

Question 7

… dependencies: dependencies over time

Answer 7

Temporal

Question 8

You will notice that in these videos I use subscripts as well as … as a numeric notation for the weight matrix.

Answer 8

superscript

Question 9

Even when knowing which annotation will get the most focus, it’s interesting to see how … softmax makes the end score become.
array:
[ 927., 397., 148., 929.]
softmax:
[ 0.11920292, 7.9471515e-232, 5.7661442e-340, 0.88079708]

Answer 9

drastic

Question 10

You could skip the embedding step, and feed in the one-hot encoded vectors directly to the recurrent layer(s). This may reduce the complexity of the model and make it easier to train, but the quality of translation may suffer as one-hot encoded vectors cannot … similarities and differences between words.

Answer 10

exploit

Question 11

If you would like to … your knowledge even more, go over the following tutorial.

Answer 11

deepen

Question 12

Words that you would expect to see more often in positive reviews – like “amazing” – have a ratio greater than 1. The more … a word is toward positive, the farther from 1 its positive-to-negative ratio will be.

Answer 12

skewed

Question 13

There are still a couple of problems to … out before we use the bigram probability dictionary to calculate the probabilities of new sentences

Answer 13

sort

Question 14

This diet is … in vitamin B.

Answer 14

deficient

Question 15

Some possible combinations may not exist in our probability dictionary but are still possible. We don’t want to multiply in a probability of 0 just because our original corpus was …

Answer 15

deficient

Question 16

Carefully read each question and provide … answers.

Answer 16

thorough

Question 17

The more … a word is toward positive, the farther from 1 its positive-to-negative ratio will be.

Answer 17

skewed

Question 18

You had to calculate all derivatives for all your functions, apply the chain rule, and then implement the result of the calculations, praying that everything … … right.

Answer 18

was done