![Venelin Valkov](/img/default-banner.jpg)
- Видео 143
- Просмотров 1 835 060
Venelin Valkov
Болгария
Добавлен 6 авг 2011
Misusings of AI, Stats & Programming
AI Agents with LangGraph & Llama 3 | Control the Execution Flow and State of Your Agent Apps
LangGraph allows you to create complex, looping workflows (instead of just linear steps) and gives you full control over how your application runs, and saves its progress. Let's learn how to build a simple app that controls our habit tracker with a set of custom tools.
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/langgraph-basics
LangGraph home page: langchain-ai.github.io/langgraph/
Follow me on X: venelin_valkov
AI Bootcamp: www.mlexpert.io/bootcamp
Discord: discord.gg/UaNPxVD6tv
Subscribe: bit.ly/venelin-subscribe
GitHub repository: github.com/curiousily/AI-Bootcamp
👍 Don't Forget to Like, Comment, and Subscribe for More Tutorials!
00:00 - Intro
00:32 - Text t...
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/langgraph-basics
LangGraph home page: langchain-ai.github.io/langgraph/
Follow me on X: venelin_valkov
AI Bootcamp: www.mlexpert.io/bootcamp
Discord: discord.gg/UaNPxVD6tv
Subscribe: bit.ly/venelin-subscribe
GitHub repository: github.com/curiousily/AI-Bootcamp
👍 Don't Forget to Like, Comment, and Subscribe for More Tutorials!
00:00 - Intro
00:32 - Text t...
Просмотров: 695
Видео
RAG from Scratch with Llama 3.1 | Build Chatbot with Custom Data, Groq API, Sqlite-vec and FastEmbed
Просмотров 2 тыс.День назад
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/rag-from-scratch Learn how to build a simple RAG system without external libraries like LangChain and LlamaIndex Follow me on X: venelin_valkov AI Bootcamp: www.mlexpert.io/bootcamp Discord: discord.gg/UaNPxVD6tv Subscribe: bit.ly/venelin-subscribe GitHub repository: github.com/curiousily/AI-Bootcamp 👍 Don't Forget...
LLM Function Calling (Tool Use) with Llama 3 | Tool Choice, Argument Mapping, Groq Llama 3 Tool Use
Просмотров 1,8 тыс.14 дней назад
What happens when you give the power of function calling to your LLM? In this video, you'll learn how to "teach" Llama 3 to understand your application by providing a set of functions to operate a complete habit-tracking app. Llama 3 Groq Tool Use: wow.groq.com/introducing-llama-3-groq-tool-use-models/ Berkeley Function-Calling Leaderboard: gorilla.cs.berkeley.edu/leaderboard.html Llama 3 Groq ...
Local RAG with Llama 3.1 for PDFs | Private Chat with Your Documents using LangChain & Streamlit
Просмотров 7 тыс.21 день назад
Learn how to build a completely local RAG for efficient and accurate document processing using Large Language Models (LLMs). Learn how to: - Extract high-quality text from PDFs using pypdfium2 - Split and format documents for optimal LLM performance - Create and configure vector stores with Qdrant - Implement advanced retrieval techniques with FlashrankRerank and LLMChainFilter - Seamlessly int...
Fine-Tuning Llama 3 on a Custom Dataset: Training LLM for a RAG Q&A Use Case on a Single GPU
Просмотров 11 тыс.Месяц назад
Are you happy with your Large Language Model (LLM) performance on a specific task? If not, fine-tuning might be the answer. Even a simpler, smaller model can outperform a larger one if it's fine-tuned correctly for a specific task. In this video, you'll learn how to fine-tune Llama 3 on a custom dataset. Model on HF: huggingface.co/curiousily/Llama-3-8B-Instruct-Finance-RAG Philipp Schmid Post:...
SQL AI Agents: Analyze Relational Databases with Natural Language using Llama 3 (LLM) and CrewAI
Просмотров 3,9 тыс.Месяц назад
Imagine giving your AI the power to work directly with your private data. That's what happens when you connect agents to a SQL database. In this video, you'll learn how to use Llama 3 with CrewAI to build a team of agents that analyze data from a database. Follow me on X: venelin_valkov AI Bootcamp: www.mlexpert.io/bootcamp Discord: discord.gg/UaNPxVD6tv Subscribe: bit.ly/venelin-su...
DeepSeek Coder v2: First Open Coding Model that Beats GPT-4 Turbo?
Просмотров 2,2 тыс.2 месяца назад
Coder v2 by DeepSeek is a Mixture-of-Experts LLM fine-tuned for coding (and math) tasks. The authors say it beats GPT-4 Turbo, Claude3-Opus, and Gemini 1.5 Pro. It comes in two versions - 16B and 236B and supports 128k context length. Let's find out how good it really is! Playground: chat.deepseek.com/coder HuggingFace Model: huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Instruct Dataset: huggin...
LLM JSON Output - Get Valid JSON with Pydantic and LangChain Output Parsers
Просмотров 1,4 тыс.2 месяца назад
How do you get a valid JSON output of Open LLMs (even when they don't support it?). In this video, you'll learn how to parse responses with prompting, Pydantic objects and/or using LangChain. Follow me on X: venelin_valkov AI Bootcamp: www.mlexpert.io/bootcamp Discord: discord.gg/UaNPxVD6tv Subscribe: bit.ly/venelin-subscribe GitHub repository: github.com/curiousily/AI-Bootcamp 👍 Do...
Situational Awareness: From GPT-4 to AGI | Compute, Algorithms & Unhobbling by OpenAI Ex-Employee
Просмотров 1,2 тыс.2 месяца назад
"Look. The models, they just want to learn. You have to understand this. The models, they just want to learn." - Ilya Sutskever (circa 2015, via Dario Amodei) Leopold Aschenbrenner is an ex-OpenAI employee who gives his thoughts on what is happening with AGI and when it is happening in his essay. Let's have a look! - Situational Awareness: The Decade Ahead - situational-awareness.ai/ - Leopold ...
Evaluate LLM Systems & RAGs: Choose the Best LLM Using Automatic Metrics on Your Dataset
Просмотров 1,3 тыс.2 месяца назад
Learn how to effectively evaluate new Large Language Models (LLMs) using automated metrics on custom datasets. Learn the best practices for choosing the right LLM for your specific project and see how they perform on various tasks. Follow me on X: venelin_valkov AI Bootcamp: www.mlexpert.io/bootcamp Discord: discord.gg/UaNPxVD6tv Subscribe: bit.ly/venelin-subscribe GitHub repository...
AI Agents with LangChain, CrewAI and Llama 3: Build an AI Tweet Writing App | Step-by-Step Tutorial
Просмотров 2,4 тыс.2 месяца назад
Learn how to build a cutting-edge AI tweet writing app using AI Agents and Llama 3. In this video, you will use CrewAI to create agents that do web scraping, researching, writing, and editing tweets. TweetCrafter on GitHub: github.com/curiousily/tweetcrafter/ Follow me on X: venelin_valkov AI Bootcamp: www.mlexpert.io/bootcamp Discord: discord.gg/UaNPxVD6tv Subscribe: bit.ly/venelin...
Can GPT-4o's Memory Replace RAG Systems? Exploring Large Context Windows
Просмотров 2,6 тыс.2 месяца назад
Large Language Models (LLMs) have struggled with large contexts (especially in the middle of the prompt: arxiv.org/abs/2307.03172 ), but GPT-4o is changing the game. Discover how it outperforms previous models on needle-in-haystack-like benchmarks, finding specific texts in vast documents. Could this breakthrough make Retrieval-Augmented Generation (RAG) systems obsolete? Let's find out! Needle...
GPT-4o API Deep Dive: Text Generation, Streaming, Vision, and Function Calling
Просмотров 2 тыс.3 месяца назад
Welcome to an in-depth look at the GPT-4o (via the API) by OpenAI! This model is currently ranked #1 on the LMSYS Chatbot Arena Leaderboard. It excels in multimodal tasks, handling both text and images effortlessly, and is highly proficient in function calling. In this video, we'll explore the capabilities of GPT-4o through various tasks, demonstrating its inference speed and performance. Learn...
GPT-4o - LMM (Audio, Vision & Text) by OpenAI | Faster, Cheaper & Smarter than GPT-4 Turbo
Просмотров 8753 месяца назад
Meet GPT-4o (omni), OpenAI's advanced Large Multimodal Model (LMM). This powerful AI can take in text, audio, and images, and generate text, audio, and images in response. It performs just as well as GPT-4 Turbo when handling text in English and code, and it's even better with non-English languages. Plus, it's much faster and costs 50% less to use through the API. Blog Post: openai.com/index/he...
Advanced RAG with Llama 3 in Langchain | Chat with PDF using Free Embeddings, Reranker & LlamaParse
Просмотров 14 тыс.3 месяца назад
Let's build an advanced Retrieval-Augmented Generation (RAG) system with LangChain! You'll learn how to "teach" a Large Language Model (Llama 3) to read a complex PDF document and intelligently answer questions about it. We'll simplify the process by breaking the document into small pieces, converting these into vectors, and organizing them for fast answers. We'll build our RAG using only open ...
CrewAI with Open LLM (Llama 3) using Groq API: AI Agents for Data Analysis with Custom Tools
Просмотров 7 тыс.3 месяца назад
CrewAI with Open LLM (Llama 3) using Groq API: AI Agents for Data Analysis with Custom Tools
AI Agents with GPT-4 Turbo and CrewAI | Cryptocurrency Market Report with News
Просмотров 8413 месяца назад
AI Agents with GPT-4 Turbo and CrewAI | Cryptocurrency Market Report with News
Run Your Own AI (Mixtral) on Your Machine - Inference using Llamacpp on a Cloud GPU (Runpod)
Просмотров 7174 месяца назад
Run Your Own AI (Mixtral) on Your Machine - Inference using Llamacpp on a Cloud GPU (Runpod)
Build Real-World Machine Learning Project: Step-by-Step Guide using FastAPI, DVC & Poetry
Просмотров 1,1 тыс.4 месяца назад
Build Real-World Machine Learning Project: Step-by-Step Guide using FastAPI, DVC & Poetry
Grok-1 Open Source: 314B Mixture-of-Experts Model by xAI | Blog post, GitHub/Source Code
Просмотров 6 тыс.5 месяцев назад
Grok-1 Open Source: 314B Mixture-of-Experts Model by xAI | Blog post, GitHub/Source Code
Real-World PyTorch: From Zero to Hero in Deep Learning & LLMs | Tensors, Operations, Model Training
Просмотров 2,2 тыс.5 месяцев назад
Real-World PyTorch: From Zero to Hero in Deep Learning & LLMs | Tensors, Operations, Model Training
Will AI Take Your Job? Should You Learn Programming and AI/ML Development in 2024 and beyond?
Просмотров 8335 месяцев назад
Will AI Take Your Job? Should You Learn Programming and AI/ML Development in 2024 and beyond?
Deploy (Tiny) LLM to Production: Merge Lora Adapter, Push to HF Hub, Rest API with FastAPI & Docker
Просмотров 1,5 тыс.5 месяцев назад
Deploy (Tiny) LLM to Production: Merge Lora Adapter, Push to HF Hub, Rest API with FastAPI & Docker
Fine-tuning Tiny LLM on Your Data | Sentiment Analysis with TinyLlama and LoRA on a Single GPU
Просмотров 14 тыс.6 месяцев назад
Fine-tuning Tiny LLM on Your Data | Sentiment Analysis with TinyLlama and LoRA on a Single GPU
Mamba vs. Transformers: The Future of LLMs? | Paper Overview & Google Colab Code & Mamba Chat
Просмотров 6 тыс.7 месяцев назад
Mamba vs. Transformers: The Future of LLMs? | Paper Overview & Google Colab Code & Mamba Chat
Key Principles for Optimizing LLaMA 2 & ChatGPT Responses | Mastering AI Prompt Engineering
Просмотров 9867 месяцев назад
Key Principles for Optimizing LLaMA 2 & ChatGPT Responses | Mastering AI Prompt Engineering
Phi 2: Small Language Model Better Than 7B LLMs? | Google Colab Tutorial with Python
Просмотров 4,1 тыс.7 месяцев назад
Phi 2: Small Language Model Better Than 7B LLMs? | Google Colab Tutorial with Python
Mixtral - Mixture of Experts (MoE) Free LLM that Rivals ChatGPT (3.5) by Mistral | Overview & Demo
Просмотров 2,6 тыс.8 месяцев назад
Mixtral - Mixture of Experts (MoE) Free LLM that Rivals ChatGPT (3.5) by Mistral | Overview & Demo
Google Gemini - Can it Beat ChatGPT/GPT4? | Technical Report - Model Architecture, Dataset, Training
Просмотров 6298 месяцев назад
Google Gemini - Can it Beat ChatGPT/GPT4? | Technical Report - Model Architecture, Dataset, Training
SDXL Turbo - (Almost) Instant High-Quality Images with One-Step Text-to-Image AI Model
Просмотров 1,1 тыс.8 месяцев назад
SDXL Turbo - (Almost) Instant High-Quality Images with One-Step Text-to-Image AI Model
About half way through, video is excellent, nicely explained, very clearly done with good pace. No music is such a treat!
Hello, thanks for the great tutorial. I'm stuck at this section: "db = sqlite3.connect("example.db") db.enable_load_extension(True) sqlite_vec.load(db) db.enable_load_extension(False)" I'm doing everything but it says "OperationalError: The specified module could not be found." on sqlite__vec.load() side. Can you help me on that?
Awesome video <3 Thank bro <3
Awesome content. Request you to categorise your channel videos into ordered playlists
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/langgraph-basics
Can I export the chatbot so I can integrate it into an app?
can you share the notebook?
Hello Venelin, thanks for the video. Iface this problem when running the query: "I encountered an error while trying to use the tool. This was the error: StructuredTool._run() missing 1 required keyword-only argument: 'config'. Tool list_tables accepts these inputs: list_tables() - List the available tables in the database" any help here? Thanks
I’m trying to fine-tune the LLaMA 3.1 8B model without quantization, but when I try to use the pipeline with the unquantized model, I encounter this error: Trying to set a tensor of shape torch.Size([128256, 4096]) in 'weight' (which has shape torch.Size([128264, 4096])), this looks incorrect. Do you know why this is happening? maybe i'm using the wrong pad token idk. Thanks in advance
Amazing video
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/rag-from-scratch
Hi, how's it going? I want to integrate a chatbot into my website. NO WhatsApp. Is it possible to create a CMS to manage this chatbot?
Possible - yes! It will be a project though, wouldn't go with building it from scratch. Checkout LangChain and/or LlamaIndex as starting points. Look at pgvector if you happen to use Postgres/Supabase. Also, there exist some premade solutions that work quite well. Btw, me and a few friends are building exactly such system that will be made available to the public, will make a video once it is released. Best of luck!
Great tutorial as usual😊. I have two queries: 1. You are using sqlite-vec for fast retrieval? Otherwise just store query and response in any db and using cosine similarity on the customer query fetch the top most relevant response. 2. Why are you using LLMs here, is it just bcoz to get the concise and short answer? Because LLMs might hallucinate in some cases which i think can be more harmful sometimes.
1. I use it because it is fast and easy to understand for this use case (simple RAG from scratch). You can use whichever vector store or database works for you. I prefer SQLite because it is an easy "run your database everywhere" solution. sqlite-vec works with SQLite and performs the similarity search for you. One general rule I follow is, "When the database can do something for you, let it do it" (as long as it is fast and simple enough). A great benefit of this approach is that your code often becomes simpler to read and understand. 2. The LLM is responsible for combining the context (retrieved questions and answers) and providing an understandable and relevant answer to the user's query. I completely agree with you that the LLM can sometimes produce incorrect responses, but it generally provides more conversational answers and takes into account the chat history. Thank you for your kind words!
@@venelin_valkov thank you so much for your kind response.
if we want retrieve the images related to the text or the question will it be possible if yes how
Same question
Hi.. how can i subscribe to ml expert
Hi, You can subscribe for MLExpert here: www.mlexpert.io/ Let me know if you have any questions! Thank you for watching!
If we apply the same approach to a dataset with another language like France or Arabic, will the approach change? Assume same columns structure, and names like you showed in the video.
thankyou so much for the vids.... Amazing content... just one query Shouldnt top be max(ys)?? at 24:27??
That's one more insightful tutorial. Thank you Venelin!
If you find a function that can completely parse any given python function into a valid tools entry, tell me :)
Can llm be used for controlling robots ?
Keep the good work
Great videos
Hi
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/llm-function-calling
So i guess this is another tutorial that ends after writing the basic thing with a "in the next video we'll make it do something" but that video never came.. I just want a single full start to end tutorial of a neural network....
can we follow the same steps to finetuning another model LMM like Mistral AI ?
is there any method to correct that mistake about calculating profits?
Nice. What I am looking for though is a similar system for not just one document but one that is suitable to shat with a corpus of 1000s of documents in my archive, and - to be greedy - to create a relational map from the perspective of the query. Any hints?
👂
Is it safe to use huggingfaceembeddings for confidential data? How secure it is?
anyoneelse have the issue with loading dataset????
ruclips.net/video/3crJNnAXd6w/видео.html
How do you modify the code so that it starts with the same document you already ingested when you restart the app?
Hi Thank you for sharing your knowledge. Please share the code of this video if possible. Thank you.
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/ragbase-local-rag
Thank you in have watch several videos on llama fine tuning With lots of differences you are the best ! Especially anout dataset and how you formatted it
I'm making my final year project and it involves training a model on custom data, thanks for this video.
Дали base model-a ще може да се тренира с двойки въпрос-отговор на Български?
when i run trainer.fit(model,data_module) 54 has_train_dataloader = is_overridden('train_dataloader', model) 55 if not has_train_dataloader: ---> 56 raise MisconfigurationException( 57 'No `train_dataloader()` method defined. Lightning `Trainer` expects as minimum a' 58 ' `training_step()`, `train_dataloader()` and `configure_optimizers()` to be defined.' MisconfigurationException: No `train_dataloader()` method defined. Lightning `Trainer` expects as minimum a `training_step()`, `train_dataloader()` and `configure_optimizers()` to be defined how to fix
Sir could you share the text summarisation with N-grams using deep learning back end code 😢
I trained it with 2 epochs and the result was amazing! Nice explanation btw!!
Could you share your use case?
that's great; llama-parse is sweet. Please make a video showing how to use a knowledge graph index in conjunction with the vector DB!
If you don't know what are you doing then don't make videos !! 🤬
Great breakdown. Keep it up
nice wutang shirt and great content. thanks
Can we fine tune to 2bit model ?
Great video! Could you share how to build the webpage with FastAPI like you did? or any tutorials, thanks!
Im looking for this ❣️
Great stuff as usual. Very useful info!
Full-text tutorial (requires MLExpert Pro): www.mlexpert.io/bootcamp/fine-tuning-llama-3-llm-for-rag What performance did you get with your fine-tuned model?
How to buy monthly subscription please let me know any link for it? As the link says yearly need the link for monthly