Encoder-decoder architecture: Overview

preview_player
Показать описание
The encoder-decoder architecture is a powerful and prevalent machine learning architecture for sequence-to-sequence tasks such as machine translation, text summarization, and question answering. Watch this video to learn about the main components of the encoder-decoder architecture and how to train and serve these models.

Enroll on Google Cloud Skills Boost to view the lab walkthrough and participate in a hands-on lab!

Рекомендации по теме
Комментарии
Автор

I came to the comments to see if I was the only one struggling with this explanation. This was a very difficult video to follow, anyone have a better way of summing this up?

DogBlessGeesus
Автор

Nothing understood. What is happening ? A simple transformer block itself contains Encoder and Decoder. Why there is explanation with RNN ? Last it says, RNNs are replaced by transformers. Makes me more confused !

abdulanzil
Автор

Anyone has link to an easier explanation? I could not decode this lesson.

debadri
Автор

You need to use encoder decoder to understand this course

gkay
Автор

I think there's definitely a simpler way to explain how this architecture works, this one is too hard to understand tbh

ThisIsAnInactiveChannel
Автор

Dam, sounds like chamber in valorant game

alaasalamah
Автор

"shifted to the left"...? 5:01

hlodwig_
Автор

enunciation and audio are so suppressed.

aashishrulz
Автор

I think this video omits too many details to the point where I don't see what was the point of recording this? The video doesn't even show how data is embedded into the encoder and how the vector is produced

NormalHuman
Автор

what am i meant to understand from this??

akheelkale
Автор

Far too complicated and difficult to understand.

Neil