Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

I recorded this introduction to embeddings (perhaps the most essential concept behind how LLMs.) But this is different from everything else you've seen: We will generate embeddings from scratch using a Siamese Network and a contrastive loss. This video is for 5 year olds.

Santiago

451,613 subscribers

68,976 просмотров • 2 лет назад •via X (Twitter)

Наука и технологии Образование

Anya Rossi• Live Now

Private livecam show

Комментарии: 11

Фото профиля Santiago

Santiago2 лет назад

Some people say that YouTube's video interface is better. If that's you, here is the link of this video over there:

Фото профиля Santiago

Santiago2 лет назад

This is a great exercise. You can read all you want, watch videos all you want, and listen to others all you want. But nothing, *nothing* will teach you more than building things yourself. Especially when you start from first principles, and add complexity step by step.

Фото профиля Hesam

Hesam2 лет назад

I love when ML concepts are explained intuitively Bookmarked 👌🏻

Фото профиля Roshini Fernando

Roshini Fernando2 лет назад

About embeddings. Awesome 😊🔥

Фото профиля nocodedev ⚡

nocodedev ⚡2 лет назад

How to transcribe an YouTube video and ask questions about it using local llm?

Фото профиля Geoff Kim

Geoff Kim2 лет назад

This is immensely helpful. Thank you for your service. 🫡

Фото профиля teamtom

teamtom2 лет назад

thank you for the video! i tried to run your notebook in Google Colab to dig deeper but unfortunaly something was not OK. first i did not touch your code just run cell by cell but it failed with some Keras code /1

Фото профиля Santiago

Santiago2 лет назад

Not sure what's happening. I just rerun it on my computer (Keras 3.3.3). Here is the training process:

Фото профиля teck 🖖🏿

teck 🖖🏿2 лет назад

Hi, 👋 What course do you recommend for junior ?

Фото профиля Justin Borgers

Justin Borgers2 лет назад

I just have to say "Thank You" for this tutorial. You really have a talent for explaining complex concepts in a way that is easy to understand!

Фото профиля Santiago

Santiago2 лет назад

🙏🏻

Похожие видео

This is, by far, one of the best uses of modern AI. If you don't use embeddings when querying your database, you are definitely leaving a lot on the table. In this video, I'll show you how to run semantic searches using OpenAI and PostgreSQL. It's all thanks to Pgai, an open-source PostgreSQL extension: Here's what will happen: 1. We'll create a simple table with news articles 2. We'll generate embeddings for those articles 3. We'll run queries on top of those embeddings For this video, I generated the embeddings using a simple query, but pgai Vectorizer would do the same automatically as new information makes it into the database. This is awesome! If you have a PostgreSQL database with data you are searching over, you should start experimenting with semantic searches immediately. For most use cases, a combination of full-text search + semantic search is the best approach. If you don't have a PostgreSQL database around, you can try free for 30 days using Timescale: Thanks to the Timescale (now TigerData) team for partnering with me on this post!

This is, by far, one of the best uses of modern AI. If you don't use embeddings when querying your database, you are definitely leaving a lot on the table. In this video, I'll show you how to run semantic searches using OpenAI and PostgreSQL. It's all thanks to Pgai, an open-source PostgreSQL extension: Here's what will happen: 1. We'll create a simple table with news articles 2. We'll generate embeddings for those articles 3. We'll run queries on top of those embeddings For this video, I generated the embeddings using a simple query, but pgai Vectorizer would do the same automatically as new information makes it into the database. This is awesome! If you have a PostgreSQL database with data you are searching over, you should start experimenting with semantic searches immediately. For most use cases, a combination of full-text search + semantic search is the best approach. If you don't have a PostgreSQL database around, you can try free for 30 days using Timescale: Thanks to the Timescale (now TigerData) team for partnering with me on this post!

Santiago

109,517 просмотров • 1 год назад

Maglev robots from a Chinese manufacturer. Perhaps this is how the future air highways will be... When flying cars take over. This concept will be implemented and everything will be automated.🤔

Maglev robots from a Chinese manufacturer. Perhaps this is how the future air highways will be... When flying cars take over. This concept will be implemented and everything will be automated.🤔

ShanghaiPanda （ Account deactivation）

65,189 просмотров • 1 год назад

[1/n] Do distinct large models admit a simple map that aligns their embedding spaces? We show that across multimodal contrastive models—trained on different data and architectures—an orthogonal map aligns image embeddings. Strikingly, the same map also aligns text embeddings.

[1/n] Do distinct large models admit a simple map that aligns their embedding spaces? We show that across multimodal contrastive models—trained on different data and architectures—an orthogonal map aligns image embeddings. Strikingly, the same map also aligns text embeddings.

Sharut Gupta

37,180 просмотров • 4 месяцев назад

How to build an AI agent from scratch. This is the question I get the most, so I recorded this video to show you from the very beginning how you can get started. You can do this in Python, TypeScript, JavaScript, or Ruby.

How to build an AI agent from scratch. This is the question I get the most, so I recorded this video to show you from the very beginning how you can get started. You can do this in Python, TypeScript, JavaScript, or Ruby.

Santiago

71,768 просмотров • 1 год назад

Anthropic pays $750,000+ a year for engineers who can build LLMs from scratch. Not how to prompt them. Not how to fine-tune them. Not how to build RAG pipelines. But how to build them from scratch. This 2-hour Stanford lecture teaches you everything. Scaling laws. Data collection. Architecture design. Post-training alignment. Free. From Stanford. Watch first. Then read this. The lecture is the theory. And this article shows you how to actually build it (with code) ↓

Anthropic pays $750,000+ a year for engineers who can build LLMs from scratch. Not how to prompt them. Not how to fine-tune them. Not how to build RAG pipelines. But how to build them from scratch. This 2-hour Stanford lecture teaches you everything. Scaling laws. Data collection. Architecture design. Post-training alignment. Free. From Stanford. Watch first. Then read this. The lecture is the theory. And this article shows you how to actually build it (with code) ↓

Rahul

418,476 просмотров • 24 дней назад

Testing Flux.1 schnell embedding interpolation. I told claude some details about my life and had it generate 40 prompts from being born -> being dead using the data I gave it. Then interpolated between the embeddings for this video. Uses Araminta softpasty LoRA

Testing Flux.1 schnell embedding interpolation. I told claude some details about my life and had it generate 40 prompts from being born -> being dead using the data I gave it. Then interpolated between the embeddings for this video. Uses Araminta softpasty LoRA

Ostris

45,749 просмотров • 1 год назад

I feel a mess. My hard work. The tools tailored to me, bought along the way. I’ve done EVERYTHING from scratch. Who does this to people? Who steals their livelihoods? I’m devastated. From the loss and the damage. This is how I make my living. #GardeningTwitter

I feel a mess. My hard work. The tools tailored to me, bought along the way. I’ve done EVERYTHING from scratch. Who does this to people? Who steals their livelihoods? I’m devastated. From the loss and the damage. This is how I make my living. #GardeningTwitter

Dirty Garden Hoe 🌱

1,187,105 просмотров • 2 лет назад

This is not Unreal Engine 5. This is generated from scratch using AI.

This is not Unreal Engine 5. This is generated from scratch using AI.

Nicolas Neubert

174,889 просмотров • 4 месяцев назад

Perhaps this is old news and everyone else knows it, but just in case, this is a handy tip for anyone using a spline path to create geo.

Perhaps this is old news and everyone else knows it, but just in case, this is a handy tip for anyone using a spline path to create geo.

Josh Powers

14,908 просмотров • 10 месяцев назад

THIS MOD IS CODED USING SCRATCH!!!! Solomons Seared Symphony is a mod I have huge appreciation for with it's dedication of using Scratch, and with this new update, the cool art style and amazing OST NEEDS to be seen by you all!

THIS MOD IS CODED USING SCRATCH!!!! Solomons Seared Symphony is a mod I have huge appreciation for with it's dedication of using Scratch, and with this new update, the cool art style and amazing OST NEEDS to be seen by you all!

T I M E L I M E (Synecdoche Funkin ???%)

13,194 просмотров • 1 год назад

THIS GUY ONLY HAS 280 SUBS??? He designed the rifle FROM THE GROUND UP. this video will BLOW his channel up. This guy is a GENIUS. This is probably the most impressive thing I have ever seen from any video game fan.

THIS GUY ONLY HAS 280 SUBS??? He designed the rifle FROM THE GROUND UP. this video will BLOW his channel up. This guy is a GENIUS. This is probably the most impressive thing I have ever seen from any video game fan.

ringoid

1,018,582 просмотров • 2 лет назад

(1/3) insitro is excited to share our paper “EmbedGEM: A Framework to Evaluate the Utility of Embeddings for Genetic Discovery,” focusing on assessing the use of ML-derived embeddings for genetic discovery using heritability & relevance to specific diseases.

(1/3) insitro is excited to share our paper “EmbedGEM: A Framework to Evaluate the Utility of Embeddings for Genetic Discovery,” focusing on assessing the use of ML-derived embeddings for genetic discovery using heritability & relevance to specific diseases.

insitro

18,421 просмотров • 2 лет назад

This is coming from somebody that graduated college last year and uses social media as a form of news. Take it for what you Will but this is how I see it and this is the way the world is moving.

This is coming from somebody that graduated college last year and uses social media as a form of news. Take it for what you Will but this is how I see it and this is the way the world is moving.

Emma Vance

48,097 просмотров • 10 месяцев назад

Look how easy it is to build apps on $ICP 🤯 Genuinely one of the most impressive things I have came across in crypto. This short video from caffeine shows how EASY it is to make and edit apps from scratch for pretty much anything using chat 😂 The future is here DFINITY Foundation

Look how easy it is to build apps on $ICP 🤯 Genuinely one of the most impressive things I have came across in crypto. This short video from caffeine shows how EASY it is to make and edit apps from scratch for pretty much anything using chat 😂 The future is here DFINITY Foundation

ALLINCRYPTO

12,787 просмотров • 2 месяцев назад

Build a RAG Application from Scratch 🐕 This video covers the architecture behind Verba 0.3.0! The design behind Verba is to have an explicit manager for each core component of RAG pipelines (Read, Chunk, Embed, Retrieve, Generate) 1. ReaderManager: Load in your data (GitHubReader, PDFReader, SimpleReader, etc.) 2. ChunkManager: Chunk up your data into smaller bits (avoids hitting the token limit and results in better retrieval) 3. EmbeddingManager: Convert your data into embeddings (use OpenAI, Cohere, SentenceTransformer, etc.) 4. RetrieverManager: Retrieve the relevant context from your query 5. GenerationManager: Generate an answer from the retrieved chunks Test out Verba on the Weaviate AI Database docs/blogs here: Verba repo:

Build a RAG Application from Scratch 🐕 This video covers the architecture behind Verba 0.3.0! The design behind Verba is to have an explicit manager for each core component of RAG pipelines (Read, Chunk, Embed, Retrieve, Generate) 1. ReaderManager: Load in your data (GitHubReader, PDFReader, SimpleReader, etc.) 2. ChunkManager: Chunk up your data into smaller bits (avoids hitting the token limit and results in better retrieval) 3. EmbeddingManager: Convert your data into embeddings (use OpenAI, Cohere, SentenceTransformer, etc.) 4. RetrieverManager: Retrieve the relevant context from your query 5. GenerationManager: Generate an answer from the retrieved chunks Test out Verba on the Weaviate AI Database docs/blogs here: Verba repo:

Erika Shorten

155,273 просмотров • 2 лет назад

Question: - Would you be okay watching an ESPN Top 10 like this? This is how we were shown the video from behind Charlie. Stand and demand TPUSA to release the original video from behind Charlie! Candace is no different than any of us! Rise up and demand accountability!

Question: - Would you be okay watching an ESPN Top 10 like this? This is how we were shown the video from behind Charlie. Stand and demand TPUSA to release the original video from behind Charlie! Candace is no different than any of us! Rise up and demand accountability!

Blake

11,688 просмотров • 3 месяцев назад

Here is a killer way to use embeddings: I built a notebook to show you how to do anomaly detection on images. Anomaly detection is one of the most common problems in the real world. Learn how to implement anomaly detection, and you'll be able to solve a ton of use cases that happen every day: • Predicting equipment failures • Identifying defective products • Identifying tumors in medical images • Spotting counterfeit products • Detecting fraud • Flagging fake reviews • Detecting bot traffic My notebook focuses on images. It uses: • Langchain • OpenClip embeddings • Oracle vector store This is the way my example works: • You start with a collection of similar images • You can then check whether a new image "belongs" to that collection I generate embeddings and use Oracle 26ai to store them and retrieve similar images on demand.

Here is a killer way to use embeddings: I built a notebook to show you how to do anomaly detection on images. Anomaly detection is one of the most common problems in the real world. Learn how to implement anomaly detection, and you'll be able to solve a ton of use cases that happen every day: • Predicting equipment failures • Identifying defective products • Identifying tumors in medical images • Spotting counterfeit products • Detecting fraud • Flagging fake reviews • Detecting bot traffic My notebook focuses on images. It uses: • Langchain • OpenClip embeddings • Oracle vector store This is the way my example works: • You start with a collection of similar images • You can then check whether a new image "belongs" to that collection I generate embeddings and use Oracle 26ai to store them and retrieve similar images on demand.

Santiago

24,527 просмотров • 3 месяцев назад

The future of AI is open-source. And ollama is the easiest way to build AI applications with open-source LLMs. Here's how to build a free, private RAG app using open-source tools. We'll use: - Ollama for LLMs and embedding models - PostgreSQL for data storage and retrieval - pgai Vectorizer for embedding creation and sync (I use Nomic for embeddings and tinnyllama as my LLM but you can substitute them for any models on Ollama)

The future of AI is open-source. And ollama is the easiest way to build AI applications with open-source LLMs. Here's how to build a free, private RAG app using open-source tools. We'll use: - Ollama for LLMs and embedding models - PostgreSQL for data storage and retrieval - pgai Vectorizer for embedding creation and sync (I use Nomic for embeddings and tinnyllama as my LLM but you can substitute them for any models on Ollama)

Avthar

34,261 просмотров • 1 год назад

Today I’m proud to launch Context, a CLI and open-source corpus for the python ecosystem - 4M quality embeddings of the top 1218 Python Libraries. $ pip install fleet-context or download to get the entire python universe. Here’s why this is game-changing for both Devs and LLMs

Today I’m proud to launch Context, a CLI and open-source corpus for the python ecosystem - 4M quality embeddings of the top 1218 Python Libraries. $ pip install fleet-context or download to get the entire python universe. Here’s why this is game-changing for both Devs and LLMs

Nicolai Ouporov @ ICML

310,799 просмотров • 2 лет назад