Awesome content. I have a suggestion. In your bnomial series sometimes the readers haven't covered a certain topic so it'd be helpful if after giving them the feedback you could link them to a good resource that explains that concept or may link them to one of these videos. It'd be a great help.
@@underfitted I'm sorry because you already provide references in your feedback but my intention was that the reference comes from interactive places that are easy to grasp or videos such as this channel of yours where we can easily understand them. Thank you
thanks i was so confused from the articles online i did not understand what they meant by flagging data. I am opting to use a Gradient boosted Tree model and i think it has built in methods to handle missing data but is that the same as flagging that you said?
I had a survey I was working with that had a bunch of check boxes, and the data was 1 or missing. This example pretty much blows up all standard methods.
Thank you. Rarest kind of advice in ML field that I ever got (not like I have been in the field for too long anyway, still an undergrad student). I have questions though. That means that given N columns-table, the maximum number of columns possible is 2N right? Also, what if we just replace the missing values of categorical columns with a new category? Do you think the idea/intuition still works? Because I think that adding columns might increase the cost especially in a very large table with massive amounts of both row and column.
Rare advice is good. It means it makes you think :) I'm not sure I follow the idea with the 2N columns. The idea of the video is to avoid losing what could be important information: the absence of a value might be as important as the value itself.
I think I answered this on Twitter. Here is what I said there: It depends on the problem. Sometimes, the best you can do is keep the missing values. Sometimes, replacing them is a better approach. Mean/Median/Mode is just one way to approach this problem.
Your posts (Twitter + TH-cam) are more helpful than any other content for gaining intuition about data. Brief and excellent! Thank you, Santiago!
this channel will be a gem in times to come
Thank you, Ashwin! Let's see what happens. Working hard on it!
You're so brave man! For real, well done! Keep it up, we will follow !
Thank you so much!
Followed your twitter, signed up for bnomial once it was launched, and now I am in love with your channel :) Thank you for the value you are creating.
Thanks so much for the support!
Great content, clear, intuitive and to the point. Refreshing to see this kind of content not 30mins long...
Glad you liked it!
Great video. Always telling my students this and really hoping they stay aware of this in the future!
That's a great point I learned today..
Thank you man....
Definitely!
Super stuff 🔥🔥. Keep this thing rolling
Thanks 🔥
Awesome content. I have a suggestion. In your bnomial series sometimes the readers haven't covered a certain topic so it'd be helpful if after giving them the feedback you could link them to a good resource that explains that concept or may link them to one of these videos. It'd be a great help.
Great suggestion!
@@underfitted
I'm sorry because you already provide references in your feedback but my intention was that the reference comes from interactive places that are easy to grasp or videos such as this channel of yours where we can easily understand them.
Thank you
thanks i was so confused from the articles online i did not understand what they meant by flagging data. I am opting to use a Gradient boosted Tree model and i think it has built in methods to handle missing data but is that the same as flagging that you said?
Incredibly insightful. Can counting the number of unanswered questions (e. 3,0,0,1,0,2...) work too?
It definitely could! It depends on the specific problem and what information could help solve it.
How would adding another column help with missing values? please explain further
I had a survey I was working with that had a bunch of check boxes, and the data was 1 or missing. This example pretty much blows up all standard methods.
Superb insight!
Glad it was helpful!
Missing data is still data.
I really loved your channel man
Thanks
Thank you. Rarest kind of advice in ML field that I ever got (not like I have been in the field for too long anyway, still an undergrad student). I have questions though. That means that given N columns-table, the maximum number of columns possible is 2N right? Also, what if we just replace the missing values of categorical columns with a new category? Do you think the idea/intuition still works? Because I think that adding columns might increase the cost especially in a very large table with massive amounts of both row and column.
Rare advice is good. It means it makes you think :)
I'm not sure I follow the idea with the 2N columns.
The idea of the video is to avoid losing what could be important information: the absence of a value might be as important as the value itself.
Another nice video!
I think I answered this on Twitter. Here is what I said there:
It depends on the problem. Sometimes, the best you can do is keep the missing values. Sometimes, replacing them is a better approach. Mean/Median/Mode is just one way to approach this problem.
@@underfitted Yeah, I was about to edit it.
Thanks for answering 🙂
What's that keyboard? :D Btw man this content rocks. Don't stop.
Thanks man! Really appreciate the comment!
The keyboard is the MX Keys Mechanical Keyboard. The just released it.
@@underfitted how's your experience with programming on it all day? Really looking into buying it!
How much time do you spend to understand the data?
Never enough
This is soo good.
Thanks!
Awesome 😎 as always
from santiago import information
Ha ha! thank you!
How can I found you on Twitter?
@svpino
in summary: add a missing indicator
Cool!!
Thanks!
And he said his videos aren't cool 🙄🙄
I'm going to take this as a nice compliment :) Thanks!
Are you italian!?
Nope. Cuban origin
Like for the fake takes! 🤣
Ha ha, yeah... I have a ton. I enjoy looking at them, so I will keep adding them to the videos.
I hope this channel never ends and keeps spreading happiness on Datascience And Machine Learning Concepts🤍🙏🏻..GO GO SANTIAGO🌟🌟🌟🌟🌟
That's the plan! Thanks!