Unsupervised Machine Learning - Hierarchical Clustering with Mean Shift Scikit-learn and Python

Your Videos deserve to be paid mate, thanks for the free awesome sharing 

Hey, could you be talked into making a NLP oriented video? I'd be interested to see how to deal with strings as features, like POS tags and lexical features.

Amazing.

Good to hear you will be doing more about machine learning ... your videos help the beginner, like me :) ... but it sure seems like "learning" is the wrong word for this stuff.

Thanks for the vids <3

I just noticed that I may be able to use machine learning to spot anomalies in network traffic ... my day job is network security using snort, suricata, python/rails apps.  There are a lot of scholarly papers on anomaly detection, but no actual example code that I've found.  I don't know if it will work as there is a lot of network traffic even on small networks.  If anyone knows of any existing code for anomaly detection please let me know.

Thanks for the videos man, it's awesome. I've watch it all in 2 days ;). Please don't stop doing please! Don't worry about the major of the subscribers are not viewing it right now, It's a brand new territory, for most of them, doing it today, you create a name on that in the future, because no one is doing that today... If you let it now probably someone will do it. This area will be the most important development area for the next years, and you are creating your name on that for sure doing this serie. Please don't stop, please!!!! Or create some paid serie, I pay for it for sure! anyway tks mate very good didactic

I was wondering if there are some books or websites (other than scikit learn) that you could recommend about machine learning+python+scikit learn ?  I keep running into issues when expanding on the "canned" examples.

Thank you so much. You are an inspiration to me.

Your videos are of great practical value and should be compulsory viewing anyone who does machine learning. As a guy studying this stuff seriously (in grad school) you have inspired me to document what I am paying for and pass it along for free the way you have. I wish you all the best.

Hello, I know this is unrelated to this video but its kind of important and would appreciate your help. About a year ago you made a comment thanking for the 5k good and 5k bad tweets from some dataset on this video:
watch?v=ytUHvMNnzZk
I cannot access this dataset, do you still have it or could you kindly point me to where I can get similar data. Thank You, keep up the awesome videos :)

you are the best! 

The tutorial was nice and neat.

please do more! I highly enjoy these!

Thanx! this was so useful! are you planning to do a PCA tutorial?

Thank you for all the tutorials on the supervised and unsupervised ML. Looking forward to watching the project-based supervised and semi-supervised ML examples. Will appreciate the finance-oriented examples similar to the those of  supervised ML. Thank you again.

Great Video, super informative. What would be a better unsupervised learning method for sample sizes larger than 10,000?

Unsupervised Machine Learning part 3 ?? when u r going to upload.I am really interested in machine learning plz do upload as soon as possible

Great video! Do you have a tutorial as to what to use when you have more than 10000 samples?

Awesome Sentdex. Very easy to follow.

you are so intelligent dude!

Can you help me with something? how can i comunicate with you?

great videos! helped me a lot in what i'm doing for my job :) thanks !

thanks for the effort... really good videos..

Great videos. How I fit a database with 28 features and 1000 rows e that model?

Watched all of the videos and followed along. Thanks for creating this content!

This is of huge help. Thank you!

can you help point me to a demo / any material for hierarchical clustering , for text...

Thanks Sentdex for clear descriptive explainations....I request for clustering using different algorithms such as DBSCAN, Gaussian based with little bigger dataset (10,000 points) (Curse of dimensionality concept)

Great introduction of MeanShift - thanks man!

Hello I want to apply unsupervised learning to a random data I generated. Basically there are two data generated , one malicious and another honest, just binary random numbers, without label. How do I approach this problem

South Africa here...Tanks so much, you don't realise how many people you are helping. YOU ARE AN UNSUNG HERO of machine learning Jedi-Mentors

I do not see drones hanging, but I do see dog chilling in bed.

Thanks for your efforts for sharing your knowledge with us.

hahaahaha someting wong. :)

thanks a lot for your video,,, great,,,,

Ha! You got a chair!

it is possible in single class or one class classification.
k=1 cluster
but i have a three data (irisdata)

28 down. thanks for the series.

I am not getting make_blobs. How to solve this issue

Hi when I am trying to fit using "ms.fit(X)" it is giving me an error saying fit() missing 1 required positional argument: 'X'

ms.fit(X)
labels = ms.labels
cluster_centers = ms.cluster_centers_

n_clusters_ = len(np.unique(labels))

print("Number of estimated clusters", n_clusters_)

colors = 10*['r.','g.','b.','c.','k.','y.','m.']

print(colors)
print(labels)

for i in range(len(X)):
plt.plot(X[i][0], X[i][1], colors[labels[i]], makersize = 10)

plt.scatter(cluster_centers[:,0],cluster_centers[:,1],
maker = "x", s = 150, linewidths = 5, zorder = 10)

plt.show()

I got a problem
TypeError: fit() missing 1 required positional argument: 'X'