Thank you so much for this Video. This helped a Lott...Just a small question In the final table "final_pd_df" (time - 1:00:06), I see that the IDs are are shuffled? By default does it all get scattered like that? because Initially when you showed the sql ages and weights table they were ordered. What's the reason for the shuffling? I couldn't figure it out.
Thank you for this video Greg! If I may ask, what if the dataset does not show clusters on a scatterplot? what model would you use to know how many walls to build for climbers?
this is superb. in piecemeal, I know about each tool, but looking at how you use them really open my eyes and mind about the "real world". awesome video.
Great video! :D Aside from maybe having more control over the specific colors of each cluster, is there any benefit to doing the coloring of the clusters with a matplotlib scatter + iteration versus simply using seaborn and the hue parameter?
Hi Greg, NICE CLIP! I just have one question, I'v read your code on your github, I just wonder whether it's a real database or it's just created like a database. And if it 's a real database, is it possible for learners to access? Thanks!
Glad that you're looking so closely! It's technically a database as you can indeed read from it using SQL, but it's fake and made up from my own NumPy code, so I wouldn't bother using it outside the project
Perfect educational video! However, one can see from the scatter plot that the data is artificial. If the data would be approximately uniformly distributed (as it is more realistic), the k-means model would not be appropriate, right? How would you handle the model selection in that case ?
There's a lot of different clustering algorithms. Agreed, kmeans might not solve the problem. It might still be appropriate though. In higher dimensions it's harder to see what will and won't work, so you could just try various algorithms and see what is produced
Nice Video, I want to follow along but i am having trouble with the code from your github. I am a python newbie When i run your top main portion of the code to get the sql data and functions i keep getting a ValueError: Table 'Ages' already exists. Do you have any pointers on how i can get this fixed?
@@GregHogg but I don't know how to implement the Neumann boundary conditions on two or three sides of the 2D Laplace equation using finite difference method
Take my courses at mlnow.ai/!
This project will definitely motivate a real world project. Thanks Greg!
Thank you so much for this Video. This helped a Lott...Just a small question
In the final table "final_pd_df" (time - 1:00:06), I see that the IDs are are shuffled? By default does it all get scattered like that? because Initially when you showed the sql ages and weights table they were ordered. What's the reason for the shuffling? I couldn't figure it out.
Thank you for this video Greg! If I may ask, what if the dataset does not show clusters on a scatterplot? what model would you use to know how many walls to build for climbers?
this is superb. in piecemeal, I know about each tool, but looking at how you use them really open my eyes and mind about the "real world". awesome video.
Thanks! Yeah, that was the main goal of the video, glad you liked it :)
Thanks a lot for the detailed explanation man! I'm watching in parts but it's very informative! God bless you!
Thank you ❤️
Great video! :D
Aside from maybe having more control over the specific colors of each cluster, is there any benefit to doing the coloring of the clusters with a matplotlib scatter + iteration versus simply using seaborn and the hue parameter?
I haven't used Seaborn. Thank you!
Hi Greg, NICE CLIP! I just have one question, I'v read your code on your github, I just wonder whether it's a real database or it's just created like a database. And if it 's a real database, is it possible for learners to access? Thanks!
Glad that you're looking so closely! It's technically a database as you can indeed read from it using SQL, but it's fake and made up from my own NumPy code, so I wouldn't bother using it outside the project
@@GregHogg Hi Greg, Soooo nice to see your comments here! It helps, Thanks !!
@@literature8536 Of course! I reply to every comment 😃
Perfect educational video! However, one can see from the scatter plot that the data is artificial. If the data would be approximately uniformly distributed (as it is more realistic), the k-means model would not be appropriate, right? How would you handle the model selection in that case ?
There's a lot of different clustering algorithms. Agreed, kmeans might not solve the problem. It might still be appropriate though. In higher dimensions it's harder to see what will and won't work, so you could just try various algorithms and see what is produced
Nice Video, I want to follow along but i am having trouble with the code from your github. I am a python newbie
When i run your top main portion of the code to get the sql data and functions i keep getting a ValueError: Table 'Ages' already exists.
Do you have any pointers on how i can get this fixed?
Sounds like everything is working right, you're just running it again and again!
Exactly what I've been searching for.
That's amazing; glad you found it!
@@GregHogg but I don't know how to implement the Neumann boundary conditions on two or three sides of the 2D Laplace equation using finite difference method
@@ndukamoses8475 neither do I! Lol
@@GregHogg hahahaha
How do I get the notebook?
Right, thanks. I'll have it on my GitHub by tonight
@@GregHogg Thanks. Great work much appreciated
On GitHub now :) Thank you
@@GregHogg Brother Please provide the link in your description. I couldn't find the link.
Clarity !!❤️
Awesome thanks!