![Shaw Talebi](/img/default-banner.jpg)
- Видео 99
- Просмотров 1 351 604
Shaw Talebi
США
Добавлен 17 сен 2020
Data Scientist | PhD, Physics
--
I got my PhD from UT Dallas, where I applied data science techniques to human health and performance research.
During grad school, I began freelancing with my data science and research skills on Upwork and beyond.
After graduating, I worked as a data scientist for 1 year at Toyota Financial Services.
In July 2023, I left my DS role to pursue entrepreneurship full-time.
And now I am here.
--
I got my PhD from UT Dallas, where I applied data science techniques to human health and performance research.
During grad school, I began freelancing with my data science and research skills on Upwork and beyond.
After graduating, I worked as a data scientist for 1 year at Toyota Financial Services.
In July 2023, I left my DS role to pursue entrepreneurship full-time.
And now I am here.
3 AI Sales Use Cases (that are not a chatbot)
👉 CXOs, VPs, & Directors... I offer custom AI workshops: www.shawhintalebi.com/ai-workshops
Here, I share 3 ways to use AI to analyze sales data. I demonstrate each approach using real-world data from a recent warm outreach campaign.
📰 Blog: coming soon!
💻 GitHub Repo: github.com/ShawhinT/RUclips-Blog/tree/main/ai-for-business/3-sales-use-cases
--
Homepage: www.shawhintalebi.com
Socials
medium.com/@shawhin
www.linkedin.com/in/shawhintalebi/
ShawhinT
shawhintalebi
Support ❤️
www.buymeacoffee.com/shawhint
Coming up - 0:00
Introduction - 0:21
3 AI Use Cases (conceptual) - 1:40
Example Code: AI Workshop Outreach - 6:12
Use Case 1: Data Augmentation - 7:26
Use Case 2: Structuring Unst...
Here, I share 3 ways to use AI to analyze sales data. I demonstrate each approach using real-world data from a recent warm outreach campaign.
📰 Blog: coming soon!
💻 GitHub Repo: github.com/ShawhinT/RUclips-Blog/tree/main/ai-for-business/3-sales-use-cases
--
Homepage: www.shawhintalebi.com
Socials
medium.com/@shawhin
www.linkedin.com/in/shawhintalebi/
ShawhinT
shawhintalebi
Support ❤️
www.buymeacoffee.com/shawhint
Coming up - 0:00
Introduction - 0:21
3 AI Use Cases (conceptual) - 1:40
Example Code: AI Workshop Outreach - 6:12
Use Case 1: Data Augmentation - 7:26
Use Case 2: Structuring Unst...
Просмотров: 64
Видео
Local LLM Fine-tuning on Mac (M1 16GB)
Просмотров 8 тыс.14 дней назад
👉 CXOs, VPs, & Directors... I offer custom AI workshops: www.shawhintalebi.com/ai-workshops Here, I show how to fine-tune an LLM locally using an M-series Mac. The example adapts Mistral 7b to respond to YT comments in my likeness. 📰 Blog: towardsdatascience.com/local-llm-fine-tuning-on-mac-m1-16gb-f59f4f598be7?source=friends_link&sk=7c3aba481527d2b3a2eac69cd3893bab 💻 GitHub Repo: github.com/Sh...
How Much YouTube Paid Me for 1,000,000+ Views (no fluff)
Просмотров 1,1 тыс.Месяц назад
Here's how much I made in my 1st year of monetization on RUclips. This is not a flex (it wouldn't be impressive if it was 😅). Rather, it is meant to help those interested in pursuing RUclips as an income source. 🎥 RUclips Lessons: ruclips.net/video/TwVl4aA5qtk/видео.html 🚀 More on my journey: ruclips.net/p/PLz-ep5RbHosXORBcWr6dy3-Wdq9RT2n2f Book a call: calendly.com/shawhintalebi Homepage: shaw...
I Was Wrong About YouTube (what I learned)
Просмотров 985Месяц назад
Last quarter, I decided to double down on RUclips... it didn't go as expected. 👉 The Data Entrepreneurs: www.youtube.com/@TheDataEntrepreneurs 🚀 More on my journey: ruclips.net/p/PLz-ep5RbHosXORBcWr6dy3-Wdq9RT2n2f Book a call: calendly.com/shawhintalebi Homepage: shawhintalebi.com/ Socials medium.com/@shawhin www.linkedin.com/in/shawhintalebi/ ShawhinT shawhintalebi S...
3 Reasons Businesses Should NOT Use AI
Просмотров 948Месяц назад
👉 CXOs, VPs, & Directors... I offer custom AI workshops: www.shawhintalebi.com/ai-workshops Over the past year, I've talked to 65 businesses about how to leverage AI effectively. Here are the three most common problems that came up. ▶️ Series Playlist: ruclips.net/p/PLz-ep5RbHosWU-y0tppW1igv-BzSQdHmc&si=oOA5J5tUl8d9DVsv 📰 Read more: shawhin.medium.com/3-reasons-why-businesses-should-not-use-ai-...
The #1 Skill That Holds (Most) Data Scientists Back
Просмотров 7642 месяца назад
Here, I discuss why data scientists must be effective communicators and share 7 tips that have helped me improve this skill set. 📰 Read more: towardsdatascience.com/the-skill-that-holds-back-most-data-scientists-a7a7f8744080?source=friends_link&sk=656256830c405ab3fb2c12def9e72c58 Resources [1] The Storyteller’s Secret: From TED Speakers to Business Legends, Why Some Ideas Catch On and Others Do...
What Nature Can Teach Us About Business
Просмотров 4322 месяца назад
After 10 years of studying Nature, here are the 7 principles I've found most helpful as an entrepreneur. 📰 Read more: medium.com/@shawhin/how-to-build-systems-like-mother-nature-cd898a83f103 Resources [1] Biomimicry: Innovation Inspired by Nature by Janine M. Benyus [2] Antifragile: Things That Gain from Disorder by Nassim Nicholas Taleb [3] The Nature of Business: Redesigning for Resilience by...
Automating Data Pipelines with Python & GitHub Actions [Code Walkthrough]
Просмотров 3 тыс.2 месяца назад
👉 More on Full Stack Data Science: ruclips.net/p/PLz-ep5RbHosWmAt-AMK0MBgh3GeSvbCmL This is the 6th video in a series on Full Stack Data Science. Here, I use Python and GitHub actions to automate a data pipeline for FREE! More Resources: 💻 Example Code: github.com/ShawhinT/data-pipeline-example 📰 Read more: towardsdatascience.com/automating-data-pipelines-with-python-github-actions-c19e2ef9ca90...
How to Deploy ML Solutions with FastAPI, Docker, & AWS
Просмотров 6 тыс.2 месяца назад
👉 More on Full Stack Data Science: ruclips.net/p/PLz-ep5RbHosWmAt-AMK0MBgh3GeSvbCmL This is the 5th video in a series on Full Stack Data Science. Here, I walk through a simple 3-step approach for deploying machine learning solutions. More Resources: 💻 Example Code: github.com/ShawhinT/RUclips-Blog/tree/main/full-stack-data-science/ml-engineering 📰 Read more: towardsdatascience.com/how-to-deploy...
How to Build ML Solutions (w/ Python Code Walkthrough)
Просмотров 3,7 тыс.3 месяца назад
👉 More on Full Stack Data Science: ruclips.net/p/PLz-ep5RbHosWmAt-AMK0MBgh3GeSvbCmL This is the 4th video in a series on Full Stack Data Science. Here, I explain why experimentation is critical to the ML lifecycle and walk through the development of a semantic search tool for my RUclips videos. More Resources: 💻 Example Code: github.com/ShawhinT/RUclips-Blog/tree/main/full-stack-data-science/da...
How to Build Data Pipelines for ML Projects (w/ Python Code)
Просмотров 3,1 тыс.3 месяца назад
👉 More on Full Stack Data Science: ruclips.net/p/PLz-ep5RbHosWmAt-AMK0MBgh3GeSvbCmL This is the 3rd video in a series on Full Stack Data Science. Here, discuss key aspects of building data pipelines for machine learning and share Python code for pulling transcripts from all my RUclips videos. 🎥 Series Playlist: ruclips.net/p/PLz-ep5RbHosWmAt-AMK0MBgh3GeSvbCmL More Resources: 📰 Read more: toward...
How to Manage Data Science Projects
Просмотров 1,9 тыс.3 месяца назад
How to Manage Data Science Projects
4 Skills You Need to Be a Full-Stack Data Scientist
Просмотров 4,7 тыс.3 месяца назад
4 Skills You Need to Be a Full-Stack Data Scientist
How I'd Learn Data Science (if I started over)
Просмотров 2 тыс.4 месяца назад
How I'd Learn Data Science (if I started over)
I Was Wrong About AI Consulting (what I learned)
Просмотров 9 тыс.4 месяца назад
I Was Wrong About AI Consulting (what I learned)
Text Embeddings, Classification, and Semantic Search (w/ Python Code)
Просмотров 47 тыс.4 месяца назад
Text Embeddings, Classification, and Semantic Search (w/ Python Code)
How to Improve LLMs with RAG (Overview + Python Code)
Просмотров 47 тыс.4 месяца назад
How to Improve LLMs with RAG (Overview Python Code)
QLoRA-How to Fine-tune an LLM on a Single GPU (w/ Python Code)
Просмотров 55 тыс.5 месяцев назад
QLoRA-How to Fine-tune an LLM on a Single GPU (w/ Python Code)
3 Ways to Make a Custom AI Assistant | RAG, Tools, & Fine-tuning
Просмотров 22 тыс.6 месяцев назад
3 Ways to Make a Custom AI Assistant | RAG, Tools, & Fine-tuning
AI for Business: A (non-technical) introduction
Просмотров 4,2 тыс.6 месяцев назад
AI for Business: A (non-technical) introduction
5 Questions Every Data Scientist Should Hardcode into Their Brain
Просмотров 1,5 тыс.7 месяцев назад
5 Questions Every Data Scientist Should Hardcode into Their Brain
How Much YouTube Paid Me in My First 6 Months of Monetization (as a Data Science Creator)
Просмотров 2 тыс.7 месяцев назад
How Much RUclips Paid Me in My First 6 Months of Monetization (as a Data Science Creator)
4 Ways to Measure Fat Tails with Python (+ Example Code)
Просмотров 6468 месяцев назад
4 Ways to Measure Fat Tails with Python ( Example Code)
Detecting Power Laws in Real-world Data | w/ Python Code
Просмотров 1,2 тыс.8 месяцев назад
Detecting Power Laws in Real-world Data | w/ Python Code
Pareto, Power Laws, and Fat Tails-what they don’t teach you in STAT 101
Просмотров 4,1 тыс.9 месяцев назад
Pareto, Power Laws, and Fat Tails-what they don’t teach you in STAT 101
I Spent $716.46 Talking to Data Scientists on Upwork-Here’s what I learned.
Просмотров 3 тыс.9 месяцев назад
I Spent $716.46 Talking to Data Scientists on Upwork-Here’s what I learned.
I Have 90 Days to Make $10k/mo-Here's my plan
Просмотров 1,4 тыс.9 месяцев назад
I Have 90 Days to Make $10k/mo-Here's my plan
How to Build an LLM from Scratch | An Overview
Просмотров 219 тыс.10 месяцев назад
How to Build an LLM from Scratch | An Overview
Fine-tuning Large Language Models (LLMs) | w/ Example Code
Просмотров 294 тыс.10 месяцев назад
Fine-tuning Large Language Models (LLMs) | w/ Example Code
Prompt Engineering: How to Trick AI into Solving Your Problems
Просмотров 25 тыс.10 месяцев назад
Prompt Engineering: How to Trick AI into Solving Your Problems
So useful Shaw! Love seeing the breakdown in a business way 🙏🏾
Many have asked for AI use cases that are not a chatbot. Should I make more content about this?
Awesome teaching quality. Simple start into text embeddings for begineers. But it would be better to use a Open Source LLM to create embeddings. OpenAI api is for paid.
Great video
Thanks
This is some really good stuff! Thanks!
Very intersting video even for more experienced watchers
If I have EOG channels in my EEG data which already contains Eye Artefacts better then should I remove those channels before applying ICA ?
Please say in hindi.
Can i do it on mac m1 8gb ram
Shawhin, I couldn't see how you enlarged the cell size to insert your picture. Please tell me how to do that. I think I've got the rest. Thank you for the information.
Hey thanks man its really helpful
great video.
thank you for this!
can you please tell me what is bread as you're a bread enthusiast
🥖🥖🥖
Best example of multikill for freelancer working remotely is to go once in a while to a lunch with your friend in the middle of the day instead of eating on your keyboard. Benefits: - you meet a friend you havent seen for a while - your brain will rest from computer - you eat - you are refreshing knowledge about idea industry and possibly travel spots MEGAKILL
the training part is really basic. I would like to see more practical, real world preoccupations in scaling duration, synch communication costs, logging, etc.
Great suggestion. Noted for future videos :)
Awesome, thank you for the video!
easy done! great walkthrough 🙃
Very useful for me as a beginner, keep creating great tutorial videos. Thanks :)
Thank you so much Shaw! Lifesaver!
This was the best tutorial! Thank you so much for sharing your knowledge
what if you have Apple M2 Max with 96gb memory? does that mean technically there is a 96gb memory GPU?
Good question. With the M-series chips there's no CPU vs GPU memory. The important thing here is using MLX allows you to make full use of your 96GB when training models!
I ll give it a try
This is too good.
Glad you liked it :)
Thank you. that was really helpful.
Thanks Shaw. This was very useful. I shared it with some of my friends too.
Thanks for sharing :) Glad it helped!
My picture turned out really blurry once added to the excel, any idea why? (I tried several high res images) I’m not versed in excel is there’s a setting I’m missing maybe?
High res images won't work here. I used 100x100 px image. The original (high res) image looked blurry here so I took a screenshot of the area with just my face and that worked for me.
does this still work?
Why has dropout parameter to config LoRA? What drop in formular?
Hi thanks for wonderful content. Can u make a video on prompt engineering and fine tuning with code explanation for open ended QA task.
Great suggestion. I added it to the list :)
Everythings Working Great - Except for when a browser (IOS) has dark mode on, there seems to always be a big white square around the entire signature (despite having selected "No Border" on Spreadsheet). Ideas?
Try adding a border but making it colorless.
I’ve watched a few of your videos. Man, you’re really inspiring! I just got into tech about a year and a half ago as an IT analyst, and now I’m finishing up an Applied Data Science certification with Greater Learning/MIT. The entrepreneurial aspect you bring to data science is very impressive. I’ve always wondered how I can turn this skill into a business. Thank you for the vision!
Thanks for the kind words! Glad the content is helpful :) We have a community for technical folks pursuing entrepreneurship. Here's an invite in case you are interested: discord.gg/RSqZbF9ygh
I'm studying the EEG project too! hope that I can learn get some new idea from your video.
Hi Shaw!!! Really great stuff... I am loving this series!!! I echo with everyone else agreeing that your videos are super informative and hands on!!! Very Very Useful!!! Many thanks man!
Glad you like them!
Has anyone tried this on an 3.8GHz 8-core intel Core i7 chip?
MLX is specifically made for M-series chips. This example won't work with an i7.
Nice lecture, very informative! I didn't watch the video related to fat tails, but I noticed N.N. Taleb's influence, my favorite author. :-)
Same here! I actually did his summer school recently: medium.com/the-data-entrepreneurs/i-spent-2-995-on-nassim-talebs-risk-taking-course-here-s-what-i-learned-c442a55a2c64
great video! super clear and on point. Thanks Shaw!
Easy to follow and understand. Thank you such for sharing your knowledge!
thanks for your video! but how to make blog post ? how to manage the URL
Good questions. I haven't done anything as sophisticated as a blog but that is something people do. This course from GitHub might be helpful: github.com/skills/github-pages You can customize the url too. See this guide: docs.github.com/en/pages/configuring-a-custom-domain-for-your-github-pages-site
@@ShawhinTalebi Thank you for your reply! I will take a look at them
Love the video thank you for these concise tutorials! On initial inference before moving onto Fine-Tuning I can't get the generation step to produce any tokens.
Glad you like them :) Not sure what could be going wrong. Were you able to successfully install mlx_lm?
@@ShawhinTalebi I appreciate you responding, I was able to figure it out! Thank you again for the video.
Thank you so much! Could not figure it out. Your made it easy! Thank you!
Does this still work?
The API has been updated a couple times since I posted this example. See a more recent walkthrough here: ruclips.net/video/4RAvJt3fWoI/видео.html
8:20 Large Language Models Only Understands Text? They Can Recognize images and all right?
Great question. Language models only understand language (text). However, we see products like ChatGPT and Claude handle them just fine. There are two ways to do this. 1) Pass image to img-to-text then pass it to a language model 2) Create a multi-modal model (e.g. GPT-4o) which can take text, images, and audio as input
Really cool and helpful. Thank you very much. Have you perform fine-tuning in llama3.1 models successfully with this method?
I have not but it should be as easy as replacing "mistralai/Mistral-7B-Instruct-v0.2-4bit" with "mlx-community/Meta-Llama-3.1-8B-Instruct-4bit" in the example code.
Can I capture video and audio all day, with a camera in my shoulder, and finetune a model with the data every night?
Sounds like an interesting use case! This is definitely possible. Potential challenges I see are: 1) handling that much video data and 2) figuring out how to pass that data into the model (e.g. you could use a multi-modal model or find a effective way to translate it into text)
@@ShawhinTalebi some steps in between to filter the input for usability could be handy. Maybe SAM?
@@daan3298 Without knowing any details, I can imagine that being helpful. Segment with SAM then object detection with another model.
you also can fork any of these themes and build it the same way, I think it's easier to change content + you can fork any user's page from repos
Great suggestion! I'll need to try that.
Oh look, a thumbnail with someone pointing in it! What a great idea! That ought to really set you apart from the rest of the lazy, unimaginative lemming morons doing to exact same thing! 😂
ok this is SO helpful, but I really can't believe that Google has not made this simpler to do.....
They want us to suffer! XD
Thanks, great content! I really like the calm way you explain it all 👌
very very helpful