Top Tweets for #seq2seq
๐ Day 46 of my AI Journey ๐
Learned about failure modes in Encoder-Decoder & ways to fix them.
Also explored key improvements:
โ Word embeddings
โ Deep LSTMs
โ Reversing input seq
โ Attention mechanism
โ Bi-directional encoders
#DeepLearning #Seq2Seq @buildinpublic
๐ Day 45 of my AI Journey ๐
Today I learned the step-by-step working of Encoder-Decoder models! Hereโs the flow ๐งต
#DeepLearning #Seq2Seq #buildinpublic @buildinpublic
๐ Day 43 of my AI Journey ๐
Dived into Encoder-Decoder in depth:
โ Core idea & workflow
โ Encoder math for hidden states
โ Context vector & info transfer
โ Decoder math for output generation
#DeepLearning #Seq2Seq @buildinpublic #BuildInPublic
๐ Day 42 of my AI Journey ๐
Today I explored Encoder-Decoder Architecture in depth:
โ What it is & why we need it
โ Encoder: feature extraction from input seq
โ Decoder: generates target seq
#DeepLearning #Seq2Seq @buildinpublic #BuildInPublic
๐ Day 41 of my AI Journey ๐
Learned Seq2Seq Models & their challenges:
โ Encoder-Decoder for seq input/output
โ Issues: long-term deps, alignment, exposure bias, OOV words
#DeepLearning #Seq2Seq @buildinpublic #buildinpublic
๐ Day 40 of my AI Journey
Today I learned about Seq2Seq Models ๐
โ What is Seq2Seq & why we need it
โ Real-world examples:
โ Machine Translation ๐
โ Chatbots ๐ฌ
โ Text Summarization ๐
โ Key challenges:
#DeepLearning #Seq2Seq #RNN @buildinpublic #buildinpublic
โจ Day 17 of Deep Learning
Dived into Attention Mechanism in Seq2Seq ๐
๐ง How attention helps models focus on relevant parts of input
๐ Big step toward modern NLP architectures like Transformers!
#DeepLearning #AI #Attention #Seq2Seq #100DaysOfAI
Announcing the NeurIPS 2024 Test of Time Paper Awards:
https://t.co/Wyhx63A2Ek
NotebookLM is pretty good. It created a 6 min conversation of two people talking about the Sequence to Sequence paper.
#seq2seq #MachineLearning #DeepLearning #NaturalLanguageProcessing #AI #NeuralNetworks #DataScience
Is your model struggling with high latency and huge memory costs for real-time sequence processing?
๐ Introducing STAR: A transformer-based model for streaming seq2seq transduction with compression.
https://t.co/5R0wJLly5U
#NLProc #Streaming #Seq2seq #Compression #SpeechToText
A Guide to Key Terms in #GenerativeAI and #LargeLanguageModels
#LLM #ArtificialIntelligence #AI #NLP #NaturalLanguageUnderstanding #NaturalLanguageGeneration #Chatbot #Prompting #PromptEngineering #Seq2Seq #OneShot #ZeroShot #Learning
https://t.co/bX6fCCnOhK via @aster_cloud
A Guide to Key Terms in #GenerativeAI and #LargeLanguageModels
#LLM #ArtificialIntelligence #AI #NLP #NaturalLanguageUnderstanding #NaturalLanguageGeneration #Chatbot #Prompting #PromptEngineering #Seq2Seq #OneShot #ZeroShot #Learning
https://t.co/2OuWLHgmJb via @astercastercom
Itโs much more accurate to say that generative pretrained transformers (such as #ChatGPT) and latent diffusion models (such as #DALL-E, #Midjourney, & #StableDiffusion), share an #attention-based/#seq2seq heritage with #Encoder/#Decoder models for protein folding.
๐งต/๐งต
seq2seqใๅฎ่ฃ
ใใใใใงๆฌกๅ
ๆฐใไป็ตใฟใใใใตใใซใชใฃใฆใใใฎใง่ชๅ็จใซใใผใใซใพใจใใฆใฟใใใใ่ฆใใจใใชใ็พใใ
#seq2seq
#DeepLearning
Summarizers exist, and they are pretty good. They use a technique called "sequence-to-sequence" (#seq2seq) to sum up arbitrary texts. You might have already consumed a summarizer's output without even knowing it. 22/
Great, but what about #seq2seq models? We knew about T5 (small, base, and large), but recently I've discovered a set of smaller ckpts: T5-efficient family (tiny and mini). The smallest ckpt is 31M (hp)! Crazy!
https://t.co/3dn8hsUkvi
Click here [https://t.co/se0tUmXLNI] If you are interested Seq-2-Seq Models in Deep Learning, this tutorial series is for you! About 8 hours of content with Python/Keras codes for free! Enjoy!
#deeplearning #seq2seq #python #MuratKarakayaAkademi https://t.co/hBcePegIQ2
Googleใ้็บใใ้ซๆง่ฝใชใใฅใผใฉใซใใใใฏใผใฏใฎใฉใคใใฉใช FlaxใไฝฟใฃใฆใSeq2Seqใๅฎ่ฃ
ใใๆนๆณใๅฎ้ใซใใญใฐใฉใ ใๅใใใชใใๅญฆใถใใจใใงใใพใใ
https://t.co/YV46p4k5vB
#AxrossRecipe
#ๆฉๆขฐๅญฆ็ฟ
#ๆทฑๅฑคๅญฆ็ฟ
#Flax
#Seq2Seq
#ใใญใฐใฉใใณใฐๅญฆ็ฟ
#้งใๅบใใจใณใธใใขใจ็นใใใใ
Many of you asked me for a Colab Notebook ๐ to train a model to translate from SQL ๐ฝ to Natural Language ๐ฃ๏ธ (and vice versa).
Tomorrow I will release it using Google's T5 and @huggingface ecosystem.
#nlp #nlg #seq2seq #tuto
Last Seen Hashtags on Sotwe
ใใจในใฟใใฉใใญใฃใณใใผใณ
Seen from United States
SexChallenge
Seen from Egypt
momson
Seen from Turkey
AngelofBogota
Seen from Peru
monkeyappwins
Zoophilia
Seen from Turkey
2023YolundaBirizBeraberiz
Seen from United States
ุฏูููุซ_ุจุฏููู
bonnieblue
Seen from Turkey
baitedboys
Seen from Pakistan
Most Popular Users
Elon Musk 
@elonmusk
240.8M followers
Barack Obama
@barackobama
119.2M followers
Donald J. Trump
@realdonaldtrump
111.7M followers
Cristiano Ronaldo
@cristiano
111.1M followers
Narendra Modi
@narendramodi
107M followers
Rihanna
@rihanna
97.8M followers
NASA
@nasa
92.2M followers
Justin Bieber
@justinbieber
91M followers
KATY PERRY
@katyperry
87.9M followers
Taylor Swift
@taylorswift13
81.8M followers
Lady Gaga
@ladygaga
73.3M followers
Virat Kohli
@imvkohli
70.3M followers
Kim Kardashian
@kimkardashian
69.9M followers
YouTube
@youtube
68.7M followers
Bill Gates
@billgates
64M followers
Neymar Jr
@neymarjr
63.1M followers
The Ellen Show
@theellenshow
62.4M followers
CNN
@cnn
61.9M followers
Selena Gomez
@selenagomez
61M followers
X
@x
60.8M followers