your video is helpfull.you r saying that highly correlated features with the target.we no need to perform any transformation.will this impact the accuracy of the model.plz reply
Hello from Turkey! That's great content. I'd like to ask you what if our target (y) is skewed what to do then ? For example let's say I made a car price prediction model and variables are horsepower of car and color of car and car price is skewed in dataset. I did sqrt or log to car price and trained the model, got the mse and did model tuning everything is finished. And now I want to see a 180 hp black car's price. If I insert the values it will give me the sqrt'ed or log'd value right ? So if I do the reverse of log or sqrt will it give me the real car price ? Or should I do other operations ? Thank you...
Hello, if the target is skewed, you can apply log or sqrt transformation. Now the predicted value you get is not original price value, it is either sqrt or log of the price. To get the actual price from predicted value, you can apply the inverse function, for eg if transformation is square root, apply square or if transformation is log (base e) apply exponential function i.e np.exp(). I hope that helps, Thank you.
in heat map one feature has 0.002, 2nd has 0.0017.when i am going to skew, getting negative values for 2nd one -0.07723174570350672 where in first 0.2155809290498895. is it correct. why -ve values comming
Can't say that, but removing outliers will definitely reduce the magnitude of skewness, but no guarantee on totally removing the skewness. Skewness and outliers are 2 different things, A normal distribution with 0 skewness can have outliers on both the extreme ends.
removing outliers by trimming is removing records which comes in outlier i.e. you are loosing the data. removing outliers with capping is good practice and you can try, it works very well.
sir how do i find skewness of a list given as price = [14751, 16422, 15398, 9445, 12589, 11687, 10692, 8475, 11184, 9961, 12898, 11905] please do reply sir will be very helpful :-)
Thanks for the videos, where is the descriptive stats videos
statquest and khan academy videos are good.
Wow I was confused about this and you explained it so well!!!! Thank you
Today I learned so many things from you bro
your video is helpfull.you r saying that highly correlated features with the target.we no need to perform any transformation.will this impact the accuracy of the model.plz reply
Could you please share the link for Descriptive Statistics Video?
How can we remove negative and positive skewness together?
Skewness can be either positive or negative, if positive apply square root, logarithm, if negative - apply square, cube or higher powers.
Plz make video on that
Hello from Turkey! That's great content. I'd like to ask you what if our target (y) is skewed what to do then ? For example let's say I made a car price prediction model and variables are horsepower of car and color of car and car price is skewed in dataset. I did sqrt or log to car price and trained the model, got the mse and did model tuning everything is finished. And now I want to see a 180 hp black car's price. If I insert the values it will give me the sqrt'ed or log'd value right ? So if I do the reverse of log or sqrt will it give me the real car price ? Or should I do other operations ? Thank you...
Hello, if the target is skewed, you can apply log or sqrt transformation. Now the predicted value you get is not original price value, it is either sqrt or log of the price. To get the actual price from predicted value, you can apply the inverse function, for eg if transformation is square root, apply square or if transformation is log (base e) apply exponential function i.e np.exp(). I hope that helps, Thank you.
@@sukamal_das Oh now it's very clear for me. Thank you so much your contents are great.
Sincerely...
Hi can you pls share the video for discriptive statistics ?
Hi Sukamal, great explanation. For symboling, can we convert the scale to min-max(0-1) and then apply sqrt?
Hello DAS Sir,
Could you please help me or provide here a descriptive statistics video link.
Not able to find out.
Any update
Anu update
Very helpful.. Thanks
I also want to ask if we reduce skewness, is there any need to scale the data as well?
Yes scaling is still required.
Thank you.. it was indeed helpful :)
Hi. Can you please take me a paid tutorial for a case study of australian vehicle prices from kaggle? Just one session will be enough.
😀There was no skewness in the opinion of your watches. They were unanimous in that it was a GREAT video. I definitely agree. Thanks.
nice explanation sir, keep posted these kind of informative videos
in heat map one feature has 0.002, 2nd has 0.0017.when i am going to skew, getting negative values for 2nd one -0.07723174570350672 where in first 0.2155809290498895. is it correct. why -ve values comming
great video
Why skewness needs to be removed?
Instead of removing skewness, would removing outliers be enough?
Can't say that, but removing outliers will definitely reduce the magnitude of skewness, but no guarantee on totally removing the skewness. Skewness and outliers are 2 different things, A normal distribution with 0 skewness can have outliers on both the extreme ends.
removing outliers by trimming is removing records which comes in outlier i.e. you are loosing the data.
removing outliers with capping is good practice and you can try, it works very well.
sir how do i find skewness of a list given as price = [14751, 16422, 15398, 9445, 12589, 11687, 10692, 8475, 11184, 9961, 12898, 11905]
please do reply sir will be very helpful :-)
from scipy.stats import skew
import seaborn as sns
price = [14751, 16422, 15398, 9445, 12589, 11687, 10692, 8475, 11184, 9961, 12898, 11905]
# print skewness value
skew(price)
# plot distribution
sns.distplot(price)
@@sukamal_das thank you sir you are great👍🙏
hello sir how to remove the negative skewness
You can go for square/cube/ higher power transformation
can we use VIF to find the correlation i.e. multi-collinearity?
Yes you can use that too
conceptual knowledge is very good, you have earned 1 more subs. There is one just improve your playlist, rest are good.
absolutely useful video, thanks for sharing!
very useful video, thank you for sharing.
very well explained!
#name sns is not defined. what am i doing wrong here?
You are missing an import statement, add this on the top cell - import seaborn as sns
@@sukamal_das yes, I figured it out. thx anyway
subscribed, thank you!
Very very useful.
Hello I tried this code I got hist but not curveshape on it why ?
Sometimes this might happen due to different versions of matplotlib or seaborn
Tq sir....🙆♂️
I typed your code but line is not visible on histogram!! whats wrong in my code? #skewness
Sometimes this might happen due to different versions of matplotlib or seaborn
@@sukamal_das whats the solution then?
@@NehaYadav-hs1po try upgrading your seaborn version
❤ and hug. ❤❤❤❤.