t-SNE Simply Explained
Вставка
- Опубліковано 4 чер 2024
- The t-SNE method in Data Science clearly and carefully explained!
0:00 Concept of Neighbors
6:25 Neighbor Similarity
8:17 Note on Standard Deviation
10:48 Moving to Lower Dimensions
22:38 KL Divergence
Resources:
Original t-SNE Paper : lvdmaaten.github.io/publicati...
KL Divergence Video : • The KL Divergence : Da...
this guy's seriously underrated
Your channel is such a gem. I hope you get more attention, you deserve praise the likes of of 3B1B and orgo chem tutor or professor Leonard
Thanks! 😃
I don't think I've ever been explained something so well. I feel like a genius right now! thank you haha
This guy is great. Wish I had you back in 2013. For your audience that wants to practice in R.
# Load required library for t-SNE
library(Rtsne)
# Generate a random dataset with 100 data points and dimensionality of 10
set.seed(123)
n
thanks!
This channel is a gem, unfortunately found out so late. Could u also make a complete series on ML / DL / NN / Generative AI concepts with practical tutorials too?
Extremely well done! thank you for sharing this.
very clear explanation and help me understand t-sne. Great job! Subscribed your channel
Thanks for the great explanation, especially on why using Cauchy in the low-dim space!
Glad it was helpful!
I'm just flooding my brain with new information, I believe I'll understand it indue time. Nice video as always
thanks!
Awesome as usual!
Amazing explanation , thank you !!!
Hi Ritvik, could u make a video for time-series data with explanation of different seasonalities (weekend, week-days, daily, hourly etc). Main ask is to understand how to analyse one data for seeing different seasonalities, and modeling it. Moreover, a video of applying different time-series models on the same?. Thank will be very useful for people like me who have just started learning time-series dataset. Thank you very much for your efforts, and time
Beautiful lecture
Amazing Content. Please make more videos, your videos are so easy to understand. Thanks a lot
Thank you!
Wow... thank u, u really the best
This wasn't easy, but finally I got the idea. Thank you!
very good !!!
Super video! What about a video on the maths and intuition behind variational auto encoders ?❤
Great suggestion!
super!
Hi Ritvik. I believe you've glanced over the fact that when normalizing the similarities in high-dimensional space, we use just the pairs with the point in question (i), but when normalizing the similarities in the low-dimensional space, we use all pairwise points.
What is the intuition behind this?
you dint over-explain believe me! XD. Overall superbly explained, thank you very much!
What a great fucking video
its KŌ-SHE