Deep Learning 01

 0    23 fiche    mszzemanek
laste ned mp3 Skriv ut spille sjekk deg selv
 
spørsmålet svaret
accomplish
We describe how to accomplish these goals by specifying a model that represents certain beliefs
begynn å lære
osiągać
throughout
Linear algebra is a branch of mathematics that is widely used throughout science and engineering.
begynn å lære
poprzez
prerequisites
the key linear algebra prerequisites
begynn å lære
warunki wstępne
slope
the slope of the line
begynn å lære
stok / nachylenie
nachylenie linii
can be thought of
scalar can be thought of as a matrix with only a single entry
begynn å lære
można pomyśleć
to yield
We allow the addition of matrix and a vector, yielding another matrix
begynn å lære
wydawać, dawać
commutative
Matrix multiplication is not commutative
begynn å lære
przemienny
undesirable
begynn å lære
niepożądany
we can derive
begynn å lære
możemy wprowadzić
to staring
Sunset came; I stared at the México sky. Isn’t nature splendid
begynn å lære
gapić się
exclusive to
It’s not exclusive to machine learning
begynn å lære
wyłącznie dla
consecutive
Where tokens are groups of N consecutive words
begynn å lære
kolejny
discarding
and those that treat input words as a set, discarding their original order
begynn å lære
odrzucanie
discrepancy
you would face the risk of introducing small preprocessing discrepancies that would hurt the model’s accuracy.
begynn å lære
rozbieżność
reshuffling
Even within a given language, you can typically say the same thing in different ways by reshuffling the words a bit
begynn å lære
przetasowania
is an interesting one
The problem of order in natural language is an interesting one:
begynn å lære
jest ciekawa
Even further
Even further, if you fully randomize the words in a short sentence
begynn å lære
Nawet dalej / co więcej
pivotal
How to represent word order is the pivotal question
begynn å lære
kluczowy
from which spring
from which different kinds of NLP architectures spring.
begynn å lære
z którego źródła?
how/when to leverage which
Let’s see how they work, and when to leverage which.
begynn å lære
jak/kiedy wykorzystać które?
without leveraging
Meanwhile, the best score that can be achieved on this dataset without leveraging external data is around 95% test accuracy.
begynn å lære
bez wykorzystywania
How could we address this?
begynn å lære
Jak moglibyśmy się tym zająć?
sparse, sparsity
begynn å lære
rzadki, rzadkość

Du må logge inn for å legge inn en kommentar.