My man, 394 views is disheartening. Also visual methods aren't always reliable. So can you make a video about finding the optimal number of clusters using data-driven approaches?

Thank you

exactly what I needed, thanks a lot

I wish it could at least create a keyword summary window and show that in the memory window and also have an autosuggestion mode so I can select the direction I prefer it to go... and I don't mean one token a time, but... more like choose-your-own adventure games whereas we don't need to rely too much on having a memory window.

thank you so much, it finally clicked! lol

👍🏿

Great video 👍it really helped me

I don’t understand how a visual estimation of the “elbow” tells us anything mathematically. Assuming something like an exponential asymptotic approach to a “best” value, wouldn’t this elbow point apparently change with different x axis scales? Or maybe some function families have an “elbow” and others don’t. Or could it maybe be something like an inflection point is for continuous functions? Very interesting!

Great Job man!

Good insight on usage of context window.

Absolutely marvelous, well well-explained in an effortless understandable way

WOW! Thank you for teaching us about LLM and tokenization with training errors! I always wonder if we can use tokenization to teach languages. 🤓

It will not work at maximum , because for asking some specific questions about project and knowing some coding, and network architecture nuances will help you tondo that work much better and faster than for some person who doesn't know math or some coding. This will only save tour time and give more data to openai for boosting their own projects)

But i wanted to why the degrees of freedom are multiplied

You lost me when you restarted the lab, I tried to follow you but got lost at around 1:16:10, can you maybe explain that error in more detail, and the steps to fix it? It was the best tutorial I have seen up until that point. This was the code block bnb_config = BitsAndBytesConfig Thank you

I think we need to have a balance of ok/nok or 0/1 events.

Interesting

Hello , can you share me the link . Its will be very helpful . Thank You .

NameError: name 'model' is not defined

would be much better if you explained the exact steps, including background iterative calculations made, to determine each split.

can u do a little tutorial for using fiass?plssssssssssssssssssssssssssssssssssssssssssss?

you deserve a like and subscribe because you teach better than my doctor so thanks man ❤‍🔥

amazing video! Thank you!

thanks man, all those fucking indian youtubers are just reading straight from books and recording themselves, Thanks a lot for explaining from the heart!

you haven't explained how it works

Thanks!

Is the course enough to land you a job as developer, and is it up to date?

How do you use ChatGPT for… “ask ChatGPT”

This was incredibly helpful to me, such a good explanation! Thank you!

I was at a school event for my kids where people were guessing jellybeans in a jar. I started looking at the sheet and pulled out my calculator, the teacher running the game asked me what I was doing and let me take six sheets of guesses over to the side after I told him (I learned it as “wisdom of the crowd”). I put my guess down and did end up winning the prize, a donated spa basket by a local business, which my wife enjoyed. I ended up talking to a class of sixth graders about my method a couple weeks later and that ended up being the real prize, because the guy running the jellybean game was the math teacher!

I'd love to say I understood this overview, but it left me with even more questions. The biggest question is - which one is most lucrative with least effort?

Great video! The simple explanations helped me better understand this new graph

thanks for sharing

Hi, so in terms of branches of data analytics, being actuarial, business, and data visualization, which would be the best to study to start moving towards machine learning? Looking at some colleges, they require you to choose one of those branches specifically.

Great! But the colab link is "Fine-Tuning HuggingFace Models" notebook not "RAG Solutions" Notebook so please fix or give new link

Information Entrophy is only for classification trees, not regression. "Instead of using entropy, regression trees rely on variance reduction or mean squared error (MSE) to measure the "purity" or homogeneity of the data points within each node. The goal is to split the data into groups where the target values are as close to the group's mean as possible, minimizing the spread (variance) within each group."

Thank You bafana .

19691 Xzavier Crescent

Excellent video, I've been using ChatGPT it for some of these applications but this content opened my eyes for new forms of using ChatGPT. Thanks for that!

Amazing content!

Skynet is near

White Deborah Lewis Jeffrey Anderson Angela

Not helpful

Great tutor

very good lesson, thanks for what you do!

Vandervort Manor