- 34
- 993 314
Rabbitmetrics
Denmark
Приєднався 31 жов 2021
Learn the nuts and bolts of AI, Automation & Analytics with Ecommerce & Marketing use cases.
High Performance (Realtime) RAG Chains: From Basic to Advanced
I will build high-performance real-time Retrieval Augmented Generation (RAG) systems in this tutorial using Llama 3, GroqCloud, LangChain, and Redis.
The written tutorial, along with the code:
www.rabbitmetrics.com/realtime-rag-with-llama-3
Here's a tutorial on how to set up a database on Redis:
ua-cam.com/video/MXhfLUoIRno/v-deo.html
The dataset used in the video
huggingface.co/datasets/ashraq/fashion-product-images-small
▬▬▬▬▬▬ V I D E O C H A P T E R S & T I M E S T A M P S ▬▬▬▬▬▬
0:00 Intro
1:29 Installing libraries and connecting to a databases
3:25 Simple RAG Chain
5:29 Hybrid RAG Chain
7:17 Contextualized RAG Chain
The written tutorial, along with the code:
www.rabbitmetrics.com/realtime-rag-with-llama-3
Here's a tutorial on how to set up a database on Redis:
ua-cam.com/video/MXhfLUoIRno/v-deo.html
The dataset used in the video
huggingface.co/datasets/ashraq/fashion-product-images-small
▬▬▬▬▬▬ V I D E O C H A P T E R S & T I M E S T A M P S ▬▬▬▬▬▬
0:00 Intro
1:29 Installing libraries and connecting to a databases
3:25 Simple RAG Chain
5:29 Hybrid RAG Chain
7:17 Contextualized RAG Chain
Переглядів: 4 077
Відео
Data Analysis with Llama 3: Smart, Fast AND Private
Переглядів 4,7 тис.14 днів тому
How good is Llama 3 at data analysis? In this tutorial, I will set up a Langchain pipeline that dynamically utilizes Llama 3 to generate SQL queries. The written tutorial and code is available here: www.rabbitmetrics.com/smart-and-private-data-analysis-with-llama-3
How to Build a Dashboard in Minutes with LLMs
Переглядів 10 тис.21 день тому
Learn to 10X your BI development speed using LLMs with custom LangChain chains to build dashboards. Full BigQuery and Looker Studio walkthrough. Written tutorial with code used in the video: www.rabbitmetrics.com/doing-bi-with-ai How to transfer data from Shopify to BigQuery. www.rabbitmetrics.com/operationalizing-shopify-data-for-ai-bigquery-tutorial/ How to build Robust Text-to-SQL chains: ua...
Advanced SQL Generation with LangChain & Redis
Переглядів 1,7 тис.Місяць тому
In this tutorial, we'll build knowledge bases for advanced SQL generation using LangChain. Code used in the video: www.rabbitmetrics.com/building-llm-knowledge-base-for-advanced-sql-chains LangChain SQL Chain documentation python.langchain.com/docs/use_cases/sql/quickstart/ This tutorial contains a video demo of generating a service account key to connect to BigQuery. www.rabbitmetrics.com/oper...
Chatting with Your Google Analytics 4 Data: Step-by-Step Python Tutorial
Переглядів 731Місяць тому
Build your own chat interface to Google Analytics 4 data using LLMs with custom LangChain chains. Code used in the video: www.rabbitmetrics.com/chatting-with-ga4-data-using-langchain The Google Analytics 4 public dataset: developers.google.com/analytics/bigquery/web-ecommerce-demo-dataset The tutorial below contains a video demo of generating a service account key to connect to BigQuery. www.ra...
Robust Text-to-SQL With LangChain: Claude 3 vs GPT-4
Переглядів 2,1 тис.Місяць тому
Generate advanced SQL with LLMs in seconds by building custom LangChain chains. The code used in the video can be found here: www.rabbitmetrics.com/chatting-with-ecommerce-data/ ▬▬▬▬▬▬ V I D E O C H A P T E R S & T I M E S T A M P S ▬▬▬▬▬▬ 0:00 The state of Chat-to-SQL 1:22 Connecting to a database (BigQuery) with LangChain 4:10 Using out-of-the-box SQL chains 6:42 Using out-of-the-box SQL agen...
LLMs will Transform Data Science - Here's How
Переглядів 4,6 тис.4 місяці тому
LLM Function Calling will 10x your Data Science Efficiency. I'll show you how to leverage OpenAI to create a targeted communication pipeline for email marketers using the email service provider Klaviyo as a demonstration. Code and data used in the video: github.com/rabbitmetrics/openai-datascience Estimating age from name: fivethirtyeight.com/features/how-to-tell-someones-age-when-all-you-know-...
Personalizing LLMs: Step-by-Step with LangChain
Переглядів 3,8 тис.5 місяців тому
Build a LangChain LLM pipeline for Ecommerce. A step-by-step tutorial with real Shopify data for beginners. Code used in the video: github.com/rabbitmetrics/personalize-LLMs Redis QuickStart tutorial: ua-cam.com/video/MXhfLUoIRno/v-deo.html ▬▬▬▬▬▬ V I D E O C H A P T E R S & T I M E S T A M P S ▬▬▬▬▬▬ 0:00 Intro: ML pipelines vs LLM pipelines 3:04 Getting started with Redis, Feast, and BigQuery...
Personalization vs Segmentation: AI/ML Blueprint for Optimizing ROI
Переглядів 1 тис.5 місяців тому
This video will dive into the differences between segmentation and personalization. I will outline a strategy for scaling ecommerce email marketing personalization using an email service provider such as Klaviyo.
Klaviyo Meets LangChain: Feeding LLMs Customer Event Data
Переглядів 1 тис.5 місяців тому
This video is a quickstart tutorial for marketers & LLM developers interested in scaling and automating email marketing strategies with OpenAI, LangChain, and Klaviyo. Colab notebook: colab.research.google.com/drive/1DEjMRGWNPiDRYtDbT1-L6Aa1DCwvChC1?usp=sharing Setting up the Shopify development store: help.shopify.com/en/partners/dashboard/managing-stores/development-stores ▬▬▬▬▬▬ V I D E O C ...
LLM Powered Email Marketing With LangChain & Klaviyo
Переглядів 2 тис.5 місяців тому
In this video, we are going to start exploring ways you can leverage AI and machine learning to personalize your email marketing strategies Specifically, we're going to see how we combine the power LLM framework LangChain in combination with the leading email service provider Klaviyo to scale targeted communication in a way that maximizes revenue and growth. ▬▬▬▬▬▬ V I D E O C H A P T E R S & T...
Chatting with 44K Fashion Products: LangChain Opportunities and Pitfalls
Переглядів 4,3 тис.8 місяців тому
In this video, I’m going to uncover some LangChain pitfalls and opportunities by building a fashion e-commerce chatbot from scratch using generic building blocks with data from Hugging Face. Link to the code: colab.research.google.com/drive/1wLUZjt32mlBj06v7TKjEKqL2nfmfnhKb?usp=sharing ▬▬▬▬▬▬ V I D E O C H A P T E R S & T I M E S T A M P S ▬▬▬▬▬▬ 0:00 Introduction and overview 0:48 AI and Headl...
Build an Image Search Engine Using Hugging Face Libraries
Переглядів 2,8 тис.8 місяців тому
In this video, we're going to build a small Python image search application utilizing a public e-commerce product dataset. The application allows you to type in a query describing a product and will then perform a semantic search to find images that match the query. Link to the code: colab.research.google.com/drive/1lENdG00tZ-YSlHeTlDA0jzPdPI6fT4M9#scrollTo=RHzohWpg44Vc ▬▬▬▬▬▬ V I D E O C H A P...
Language Modeling with Redis (and LangChain): From Zero to Hero
Переглядів 3,2 тис.8 місяців тому
In this video, we're going to dive into the vector similarity search capabilities of Redis. I will show you how to create schemas, load vector data into Redis, write the VSS queries, and pass the data to a large language model using LangChain. Link to the Colab notebook: colab.research.google.com/drive/1qrOrdvewQXL1M-ZB3l2ygdWOcf3yMFZJ Link to the data: cseweb.ucsd.edu/~jmcauley/datasets/amazon...
Beyond Basic LLM Applications: Getting Started With Redis and LangChain
Переглядів 3,9 тис.8 місяців тому
In this video, we're going to have a closer look at how you use LangChain with Redis. I'm first going to make a case for why you want to consider Redis as a backend for your LLM application. Link to the Colab notebook: colab.research.google.com/drive/1tyils0uWUyekBLwMPHIpGVkVvpfjpPux?usp=sharing Link to the data: cseweb.ucsd.edu/~jmcauley/datasets/amazon_v2/ A full course on how to develop and ...
Cold Email Automation with LangChain, Lemlist & Apify
Переглядів 5 тис.11 місяців тому
Cold Email Automation with LangChain, Lemlist & Apify
LangChain Agents: Simply Explained!
Переглядів 55 тис.11 місяців тому
LangChain Agents: Simply Explained!
LangChain In Action: Voice of Customer Modeling With Zapier
Переглядів 11 тис.Рік тому
LangChain In Action: Voice of Customer Modeling With Zapier
LangChain In Action: Real-World Use Case With Step-by-Step Tutorial
Переглядів 66 тис.Рік тому
LangChain In Action: Real-World Use Case With Step-by-Step Tutorial
LangChain & GPT 4 For Data Analysis: The Pandas Dataframe Agent
Переглядів 55 тис.Рік тому
LangChain & GPT 4 For Data Analysis: The Pandas Dataframe Agent
LangChain Explained in 13 Minutes | QuickStart Tutorial for Beginners
Переглядів 678 тис.Рік тому
LangChain Explained in 13 Minutes | QuickStart Tutorial for Beginners
Dominate Google Analytics 4 with ChatGPT 4: This Will Transform Marketing Analytics
Переглядів 7 тис.Рік тому
Dominate Google Analytics 4 with ChatGPT 4: This Will Transform Marketing Analytics
Analyzing Data with GPT4: Are Data Analysts Doomed?
Переглядів 12 тис.Рік тому
Analyzing Data with GPT4: Are Data Analysts Doomed?
Learn to Stream Shopify Data with Python in 5 Minutes!
Переглядів 1,6 тис.Рік тому
Learn to Stream Shopify Data with Python in 5 Minutes!
Unlock the Power of Shopify Webhooks: A Python Developer's Guide
Переглядів 2,1 тис.Рік тому
Unlock the Power of Shopify Webhooks: A Python Developer's Guide
Data Wrangling With Shopify's GraphQL API: Python & Pandas QuickStart
Переглядів 2,6 тис.Рік тому
Data Wrangling With Shopify's GraphQL API: Python & Pandas QuickStart
RFM Segments In BigQuery With Google Cloud Functions: Complete Python Application
Переглядів 795Рік тому
RFM Segments In BigQuery With Google Cloud Functions: Complete Python Application
Google Cloud Functions Python QuickStart: Step-by-Step Shopify Example
Переглядів 10 тис.Рік тому
Google Cloud Functions Python QuickStart: Step-by-Step Shopify Example
Breaking Down Customer Lifetime Value: A Step-by Step Shopify Example
Переглядів 692Рік тому
Breaking Down Customer Lifetime Value: A Step-by Step Shopify Example
Feeding Google Ads with Retention & Loyalty Data: An Ecommerce (Shopify) Guide
Переглядів 495Рік тому
Feeding Google Ads with Retention & Loyalty Data: An Ecommerce (Shopify) Guide
Hi, this video is one of the best, but now langchain changed its modules and classes, please update us with the new video, for eg: simplesequentialchain is not supporting now!!
I really don’t understand the benifit of language chain. Everything it does seems useless. Prompt templates can be done with the python .format() function. I don’t see the need for pinecone… just get a list of tuples [(vector, text/data), …] and just loop over the list, that’s completely free. The agents are cool, but very simple to implement aswell, and that would be better so it could be more efficient for ur use case. The one thing I do like is the chain concept, it lets you define a architecture in a more abstract/clear way, but I don’t like the way the implemented it, it should act more like piping and IPC does because you can have branches and concurrent events.
New subscriber as you giving content a couple layers deeper than I’m seeing from others Love how you start simpler then start adding complexity and other variations
How do you manage the hallucination?
Congrats, great video! Can u share the repo of the code?
This channel knows its stuff! But if you want the best for cold emailing, Mystrika is where it is at. Their unified inbox and automatic bounce detection are unbeatable. Worth every penny!
Awesome breakdown on email automation! Have you tried Mystrika yet? It is far superior to any other tool out there. Sales and lead gen have never been easier for me since I switched.
Yo, great content as always! But if you wanna skyrocket your cold emails, Mystrika is the ultimate tool. Detailed analytics, A/B testing, and an active community of 5000 users. cannot recommend it enough!
Nice one. Whete can I get this notebook?
Thanks! There's a link with the written tutorial and the code below the video.
Wonderful video thanks!
Thank you!
Great work. Thanks for sharing. Maybe you could dive into the langchain expression language in one of your future videos. The piping and runnable is not so clear yet. br
Thanks! I agree, LCEL needs more focus. I will likely cover that in more detail as we move along.
@@rabbitmetricsyes! Please 🥺
This is one of the best asusual, thanks allot rabbit metrics, this is kept to the point
Thank you! 🙏
Build a team for fun. Have a group? 1. CodeCraft Duel: Super Agent Showdown 2. Pixel Pioneers: Super Agent AI Clash 3. Digital Duel: LLM Super Agents Battle 4. Byte Battle Royale: Dueling LLM Agents 5. AI Code Clash: Super Agent Showdown 6. CodeCraft Combat: Super Agent Edition 7. Digital Duel: Super Agent AI Battle 8. Pixel Pioneers: LLM Super Agent Showdown 9. Byte Battle Royale: Super Agent AI Combat 10. AI Code Clash: Dueling Super Agents Edition
note create_pandas_dataframe_agent has been moved to langchain experimental and you have to install if you havent
Hi I created cloud function using storage triggers, but I am not able to see print results in logs.
Content is great, but the title is very misleading. Better: self-correcting SQL-query request generator. And the specific model is less important given the self-correcting feature, which could have been more promoted instead of jumping on the llama3 hype train. Just some food for thought
best
Another video that shows how disappointing AI is. I can easily write those simple queries myself
a year ago it couldn't do that. Give it time and baby will grow.
you missed the entire point of this.
you can use the result of this chain as a dataset for fine-tuning a model.
Yes, I'm working on that. The problem is that SQL queries can be erroneous and still be executed.
Awesome tutorials and tutorial layouts. 👍
Thank you!
great video - btw, I can't download the "schema.py" file from the written tutorial
Thanks! I've updated the written tutorial and the Colab notebook with the schema extraction functions
@rabbitmetrics thank you for the great content. I'm building this pipline as a poc for a project at work. I'm running into an error when in google colab. When running 'from feast import FeatureStore' I get ImportError: cannot import name 'field_validator' from 'pydantic' . Any insights would be appreciated. I experimented with different versions of feast & pydantic to no avail.
Nice tutorial thanks. Just as a note, the link in the description is not clickable.
Thanks! It is clickable now
@@rabbitmetrics Thanks
maybe this is a dumb question, at 7:54 when you say llm=llm in that line, did you define a variable called llm somewhere ?
Wow, this video provides a fascinating insight into Langchain agents and their capabilities! 🤖 I'm excited to learn more about how agents work under the hood of Link Chain and the endless possibilities they offer for businesses investing in technology, data, and analytics. Looking forward to diving deeper into building custom agents and unlocking their potential! 🚀🔍 #LangtuneAgents #AI #Innovation
Nice, but why not use proper BI tools for this? 😂
And what do tou have in mind, pwrBI?
Will this work with the Redis alternatives now they are persona non grata?
Is it possible to connect directly to the google analytics on the platform?without the need of downloading the dataset?Thank you in advance
Phenomenal thank you 🙏🏽
Phenomenal
good instruction ...
Thanks!
Hi, great tutorial! How would you implement a chat fuctionality? where you can ask follow up questions??
Thanks! I would use ChatMessageHistory to manage the conversation and catch the traceback - this is needed for more advanced queries.
Show de bola!
Dang! all python. Maybe I'll start video using Java and LangChain4J
Wow! What a hot mess! Has it gotten any easier to do this in the past 11 months?
Hello, I just run your script around 05:53 with python3 and pip3. However it says ` Could not import openai python package. Please install it with `pip install openai`. (type=value_error)`. Which version of that dep should I add to get a coherent project with your code?
Thank you.
You're welcome
🎯 Key Takeaways for quick navigation: 00:00 *🛠️ Generación de bases de conocimiento para cadenas SQL basadas en LLM* - La creación de bases de conocimiento estructuradas permite generar consultas SQL más avanzadas. - El uso de técnicas de promoción de cortos proporciona ejemplos de consultas SQL detalladas, reduciendo errores de promoción y acelerando la construcción de productos de datos. - Se presenta una visión general de cómo funciona el proceso, desde la creación de consultas SQL relevantes hasta la generación de cadenas ejecutables basadas en un prompt. 02:17 *📊 Problemas al extraer datos anidados en BigQuery y soluciones propuestas* - Se identifica un problema común al trabajar con datos anidados en BigQuery, especialmente al utilizar el operador UNNEST. - Se muestra cómo abordar este problema mediante la construcción de cadenas SQL personalizadas y la integración de información de esquema detallada. - Se introduce el concepto de promoción de cortos y se demuestra cómo su implementación puede mejorar la robustez del proceso de generación de consultas SQL avanzadas. 05:03 *🔍 Implementación de promoción de cortos con Redis y búsqueda de similitud de vectores* - Se utiliza Redis como almacén de vectores para almacenar ejemplos de consultas SQL necesarios para la promoción de cortos. - Se describe el proceso de búsqueda de similitud de vectores para recuperar consultas relevantes basadas en un prompt específico. - Se muestra cómo utilizar un selector de ejemplos de similitud semántica para realizar la búsqueda de similitud de vectores y seleccionar ejemplos relevantes para la promoción de cortos. Made with HARPA AI
Thank you this is the info I was looking for.
Hi, question, how do you configure it to use gemini-pro and not gpt-4?
Hi, you install the LangChain integrations for Gemini pypi.org/project/langchain-google-genai/ then you import ChatGoogleGenerativeAI and define llm = ChatGoogleGenerativeAI(model="gemini-pro")
@@rabbitmetrics I did it, but it doesn't work, it has this error: TypeError: Expected a Runnable, callable or dict. Instead got an unsupported type: <class 'langchain_core.runnables.base.RunnableBinding'>
@@AndresAlarcon-bb9ql you might be passing a string instead of a function in the RunnablePassthrough?
You can use create_sql_agent right?
I prefer to build it from scratch as this allows me to control the prompting and how I fetch the schema information. GA4 data is complex and you'll likely find the create_sql_agent has difficulties dealing with repeated records and unnesting properly. Take a look at my previous video ua-cam.com/video/klHZTIzk2Hk/v-deo.html
@@rabbitmetrics that makes sense. Cool video. Gave me some ideas for my project. Thanks.
@@VijayasarathyMuthu, you're welcome. Thanks for watching
Hi, the code cannot be accessed.
Hi. I just checked, the notebook should be accessible via the link in the tutorial. Did you get access?
THIS is function-calling but instead of a "json" u get a "sql query". Am i missing something?
That is one way to think of it. But in this case LangChain is handling the parsing of the LLM output (note the "model.bind(stop=[" SQLResult:"])" in the chain). When you generate SQL or any other code you'll find that the code is often returned in quotes or with some text explaining the code. The trick is to minimize this by parsing the output in a suitable way.
What happens if he drops the table when hallucinating
Read only role
As mentioned, make sure to restrict access scope and permission.
Where can we download the code file?
There's a link below the video to the Colab notebook with code and written tutorial including how to generate the ecom tables
Good stuff on Apify, fellas! I wonder if you have crossed paths with Mystrika yet? As a user, I can vouch for their superior and comprehensive list of features. You have got to try their tag management system, it is damn neat and super convenient!
What up, Rabbitmetrics! Enjoyed your video on Email Automation. But man, nothing beats Mystrika is unique email warming-up feature for increasing deliverability. Their warmup pool is quality is top-notch and it is free for one email address! Time to upgra
Decent video on Lemlist! But man, are you missing out on Mystrika! Their inbox rotation feature alone has taken my cold email to a whole new level. The scalability is limitless! You got to do a tutorial for the community here.
Hey, good one on Automation with Lemlist! Ever tried Mystrika though? Their cold email features like dynamic email generation and the AI writer are just sublime. Plus, they have got a banging community of 5000 users on FB, youll love it!