Комментарии:
This explanation is so intuitive and amazing. Thank you very much :)
ОтветитьThat was cool 👍
ОтветитьAmazinggggg! Simple, clear and engaging content! Thank you
ОтветитьMate i just love your channel. The visual, the quality of the explainations, the shameless promotion humor haha
I just whished there were some french speaking channels as great as yours...( French being my native language).
You english is so clear that i get most of it, even if certain terms are a bit challenging for me. People would love education if most teachers were like you !
Anyway keep up the great work :)
Once we get the clusters out of algorithm, Is there any way to one level further down ? To find subclusters within clusters ?
Ответитьvery well explained
ОтветитьThat's awesome! Can you please make a video about Spectral Clustering as well?
Ответить"shameless self promotion" xD
ОтветитьSo DBSCAN works on multi-dimensional space the same way? The circles are multi-dimensional?
ОтветитьThanks❤
Ответитьwhat if the size of the point is different? what are the options to use this methodology in that case ?
ОтветитьJosh, thank you so much for all those videos. U have no idea how mych u help us out! And if I may: could u do one about the Louvain clustering?
ОтветитьReally nice explanation!
Ответитьwhat a entry 😁
Ответитьhow does DBSCAN work on multi dimensional data?
ОтветитьBam! Brain washed by that... Double Bam!
ОтветитьDat song 😆
ОтветитьNice and simple😊
ОтветитьSpeed 1.25, sometimes 1.5 feels slow with this guy
ОтветитьGood Initiative..Very well explained..
ОтветитьHave a shot everytime he says corepoint
ОтветитьHey, thanks for the amazing video! Could you make an explanation video for HDBSCAN? I'm trying this out now and it looks very promising, but I'm having a harder time figuring it out than DBSCAN.
ОтветитьTHANK YOU FOR THIS! Please also make one for Ordering Points to Identify the Clustering Structure (OPTICS)
ОтветитьI'm just binging all these classification algorithm videos, learning for my exam
ОтветитьSir, how we decide if we should use DBSCAN or KNN for a clustering problem? is there any guideline or any steps to check ?
Ответитьlike the musical call back to sorting out sorting
Ответитьhelpful
Ответитьthanks
ОтветитьThank you !!!! you're great , I love your piu-piu-piu-piu 😁
ОтветитьThanksssss!!!! and Bam!!
Ответитьyou are so smart
ОтветитьGreat Explanation. So I just have one small doubt to confirm on, since you said outliers will be left over and not added to any cluster, so the DBSCAN algorithm is not sensitive to outliers. Is this right?
ОтветитьI can't believe how many of these fancy sounding algorithms are just coloring stuff that's close to eachother the same color
Ответитьsuper nice explanation!
ОтветитьAmazing explanation Josh Sir
please make video for machine learning with python
the moment hearing the music I gave it a thumb up
ОтветитьCan't express how much I enjoy your videos. Thank you for the smile :)
ОтветитьWell I can't not subscribe to you after this.
ОтветитьPlease it would be great if you can make a video of HDBSCAN, i couldn't find any resourse which explains it without going into cokplex maths.
ОтветитьIf the dbscan data is highly imbalanced, then can we use up sampling or low sampling methods ?
Ответитьgreat explanation, thanks!
ОтветитьHow is it different from kNN?
ОтветитьSo there are 2 hyper-parameters in this algorithm : - the minimal number of neighbours to be a core point and the "radius" of a core point. That's right ?
ОтветитьExcelent!
ОтветитьThank you Joshhhh you are awesomeee!!!🥳🥳🥳🥳🥳🥳
ОтветитьIn the green cluster, why someone with a high weight low height is in the same cluster as someone with a low weight high height? I don't understand it.
ОтветитьGreat!
Ответить