Thanks for posting this course! It was super useful to get another perspective on ML concepts I learned in my degree, not to mention getting solid application guidance from then man himself!
Respected Sir, can you please provide a series of case studies over datasets from Kaggle. Generally, we can't see how top rankers have done advanced feature extraction and other things, so I request you to bring a series of 5-10 case studies through which we could learn what to apply , hot to apply, to get the best out of features, etc.
Thanks a lot for this massive load of great material to study! Even though I just finished 9 lectures so far. In the comments you mentioned reworking your book. Is there any (more or less) concrete timetable? Is there maybe other literature or lectures you'd suggest (except of those mentioned at the beginning of the lecture)? Once again: thanks a lot and greetings from Germany :)
By the end of the lecture you mention something about complex and simple models, is there any analytically/numerically way in which you can measure the complexity of a model in order to compare them ? as per the following question in stack exchange: datascience.stackexchange.com/questions/80531/is-there-any-way-to-explicitly-measure-the-complexity-of-a-machine-learning-mode Gratitude for all the knowledge you have share to the world. Regards from Mexico
There is many theoretical ways, and few practical ways. Standard measures used in machine learning theory are Vapnik-Chervonenkis dimension and Rademacher complexity. Measures used in probabilistic modeling include AIC and BIC. None of these is very useful for supervised machine learning in practice.
Professor i am unable to install fbprophet a time series forecasting tool by Facebook used in machine learning in my machine. Running into a bunch of errors. Any suggestions ?
Thanks for posting this course! It was super useful to get another perspective on ML concepts I learned in my degree, not to mention getting solid application guidance from then man himself!
Respected Sir, can you please provide a series of case studies over datasets from Kaggle.
Generally, we can't see how top rankers have done advanced feature extraction and other things, so I request you to bring a series of 5-10 case studies through which we could learn what to apply , hot to apply, to get the best out of features, etc.
A very good survey lecture especially for someone who is interested in non-traditional methods in time series.
Thanks Alot Andreas Mueller Sir For Sharing this Course , Very Grateful to You!
Thank you so much for sharing your lectures! 💯
Thanks a lot for this massive load of great material to study! Even though I just finished 9 lectures so far. In the comments you mentioned reworking your book. Is there any (more or less) concrete timetable? Is there maybe other literature or lectures you'd suggest (except of those mentioned at the beginning of the lecture)?
Once again: thanks a lot and greetings from Germany :)
is it possible that you will make more time series videos in the future?
Unlikely for the near future. This was a class at Columbia. I'll probably be working on redoing my book.
Thanks for the videos you post here! They are very helpful, I was expecting more advanced stuff... :)
By the end of the lecture you mention something about complex and simple models, is there any analytically/numerically way in which you can measure the complexity of a model in order to compare them ? as per the following question in stack exchange: datascience.stackexchange.com/questions/80531/is-there-any-way-to-explicitly-measure-the-complexity-of-a-machine-learning-mode
Gratitude for all the knowledge you have share to the world.
Regards from Mexico
There is many theoretical ways, and few practical ways. Standard measures used in machine learning theory are Vapnik-Chervonenkis dimension and Rademacher complexity. Measures used in probabilistic modeling include AIC and BIC. None of these is very useful for supervised machine learning in practice.
Professor i am unable to install fbprophet a time series forecasting tool by Facebook used in machine learning in my machine. Running into a bunch of errors. Any suggestions ?
Without the error message that's impossible to solve.