Tutorial 44-Balanced vs Imbalanced Dataset and how to handle Imbalanced Dataset
ฝัง
- เผยแพร่เมื่อ 10 ม.ค. 2019
- Here is a detailed explanation about the balanced vs imbalanced dataset and how to handle the imbalanced dataset.
#balancedvsimbalanceddataset
You can buy my book where I have provided a detailed explanation of how we can use Machine Learning, Deep Learning in Finance using python
Packt url : prod.packtpub.com/in/big-data...
Amazon url: www.amazon.com/Hands-Python-F...
![[SUB] 베몬이 교관님 멘탈을 탕탕 후루후루~٩(๑❛ᴗ❛๑)۶ FOREVER 웃게 되는 갓기들의 수련회 | 돌들의침묵 | BABYMONSTER(베이비몬스터) 'FOREVER'](http://i.ytimg.com/vi/FISAO7BwIxE/mqdefault.jpg)
You are a great tutor, the way you are explaining, great to see you holds to the subject. Awesome
Well-done sir. Thanks for sharing very good to understand everyone
Great information,, good explanation.. 👌👌
Can u show the techniques of handling missing data excluding mean,median mode technique. I read that u can use regression or classification technique to find missing values but never seen implemented.
your teaching is very effective sir.... very easy to understand
great video, easy to understand.
Very nice explain, Thanks a lot.
Great sir there is no need to join any institute to learn the Data Science just follow the Krish Naik sir playlist.
Very informative, thank you.
Thanks Krish
XGboost will take care of the weights by itself and we don't need to do any weight adjustment manually ... is that right??
Great videos .. thanks for sharing them .. I am sure you will get big number of followers very soon!!
I am also applying XGBoost to the DataScience youtube channels that I am following and your channel seems to be getting heavier and heavier weights ;)
while doing downsampling as mentioned we will be reducing the points to 100 so which points to be reduced or simply picking in random is also not a good practice so how could we select those 100 points.
Simplistic approach for Beginners, would really appreciate if you could do demo with dataset(1000 rows ) of 4-5 features, as well please explain Xgboost. Thanks in advance. Look forward to you response.
Nice series. Get going. The numbers should be 630, 270 and the accuracy exactly 90%. This is the same as your original imbalance.
Can you make video/tutorial on hyper parameter tuning in classification algorithms!! As it has been coming to many interviews also.
good explanation..Thanks..
sir suppose if i have less data to train my model at the time what i have to do?
Thanks Sir, plz make a video for overfitting and underfitting
Really helpful 🙏
Very helpful video
Thank you so much
Can we do down sampling in credit card fraud detection dataset?
can u please explain the that 2 algorithms xgboost and adaboost?
how to handle missing data if data is 1tb or more? please explain
Upto what ratio should we consider it as a balanced dataset ??
@Kushal Hu what ratio that is?
Sir, I have data with binary classification output variables. the ratio of classes is 7.5:2.5. Is this balanced data or unbalanced data?
very useful . thanks. could you pls post video on PCA and LDA and about Regularization as well. thanks
Thanks, please check my playlist, video is already there for PCA
How we can downsample data points ?
Sir can you please do this same as practically. Thank you sir
how to deal with imbalance dataset when we have multiclass in target variable?
use other metrics such as F1 score instead of accuracy...
Hi krish
Plz upload vedios on interview questions
From which playlist this video is?
I am just asking,
So, if we use xgboost algorithm for classification problem we no need to bother about class imbalance.
Am I right sir.
Thankyou Sir
great content. would be nice if it goes deeper
Sir I have a doubt in this question that how to handle imbalanced data-set either by using right evaluation metrics or by these sampling techniques???Please let me know I'm very much confused...
aayushi jain SMOTE can be used where we are increasing the minority data. Safest technique
Krish : Can you put this in MACHINE LEARNING playlist? If I am not missing anything
can you explian with python codes ?
any videos for continuing these
Can you share python code to implement these up sampling techniques. Is this concept is also applicable to NLP datasets?
If working with tf.keras, add following code to model:
1) from skearn.utils import class_weight
2) class_weights = class_weight.compute_class_weight('balanced',np.unique(y_train),y_train)
3) history = tf.keras.model.fit(x_train, y_train, batch_size=x, class_weight=class_weights)
Discuss everything in upcoming video..? :p
Waiting for more interview questions
You can find the complete playlist on the below youtube url
th-cam.com/play/PLZoTAELRMXVPkl7oRvzyNnyj1HS4wt2K-.html
I will be updating this with all the questions
i really happy with imbalanced and balanced dataset explaination.......
how is accuracy 350/30 ?
got it :) 350/380
Small doubt missclassfication and imbalanced are same or different..
Pls answer me
reply this question please
It should be 250+ in test not 350+
Consider using better whiteboard software. Your written words and numbers are nearly impossible to read in Paint.
Hi Davinder, feedback taken
what i feel when see you videos explanation, your explanation like as when you communicating in english in your video its feels like you are communicating in regional language ..........thats why i understand more.........you explain most of the difficult terms in simple way. I love your explanation.
and also i remembered most of the concept in long time .
and i also like your videos which i watched.
You know..? :p