BLEU Score Explained

preview_player
Показать описание
In this video, we talk about the BLEU (BiLingual Evaluation Understudy) score which is a popular metric for evaluating machine-translated text, or text generation models, in general.

*Related Videos*
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

*Contents*
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
00:00 - Intro
00:19 - Precision and recall
00:51 - Precision and recall in language
01:45 - Precision and recall w/ n-grams
02:27 - BLEU score
03:58 - Brevity penalty
05:32 - Outro

*Follow Me*
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

*Channel Support*
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
The best way to support the channel is to share the content. ;)

If you'd like to also support the channel financially, donating the price of a coffee is always warmly welcomed! (completely optional and voluntary)
► Bitcoin (BTC): 3C6Pkzyb5CjAUYrJxmpCaaNPVRgRVxxyTq
► Ethereum (ETH): 0x9Ac4eB94386C3e02b96599C05B7a8C71773c9281
► Cardano (ADA): addr1v95rfxlslfzkvd8sr3exkh7st4qmgj4ywf5zcaxgqgdyunsj5juw5
► Tether (USDT): 0xeC261d9b2EE4B6997a6a424067af165BAA4afE1a

#bleu #machinetranslation #nlp
Рекомендации по теме
Комментарии
Автор

thanks a lot! please keep making more videos. they are very useful and informative. videos like yours make learning so much fun and engaging :D

Accorinrin
Автор

I just love you so much! I could really see your effort to make things intuitive. You are so underrated and I say this from the bottom of my heart.

stanleykurniawan
Автор

recall doesn't worry about FP. why is it 2/5? If we have all ground truths in a given list of predicted positives, isn't recall 1?

lamrin