Performance Metrics On MultiClass Classification Problems

preview_player
Показать описание
Please join as a member in my channel to get additional benefits like materials in Data Science, live streaming for Members and many more

Please do subscribe my other channel too

Connect with me here:

Рекомендации по теме
Комментарии
Автор

Guys, lets applaud his efforts. This video is being shot at 12 in the morning :)

vibesnovibes
Автор

To the point video. Awesome Krish. Your videos save hours of efforts for many students.

ashishmondal
Автор

I follow ur machine learning videos & it helps me a lot. Could you please make a video on multilevel multiclass performance metrics (absolute-true, absolute-false, accuracy, coverage, aiming)? I can't find out any appropriate source for it! Please try to do that!

afrida
Автор

That was a superb video.Got all my doubts clear.If you could the ipynb notebook or google colab for practise would be great .Thanks

sandipansarkar
Автор

@Krish Naik. If we have too many classes, then I think confusion matrix will create confusion and not be that much easy to interpret. Sir any other performance matrix for multiclass classification except confusion

AjaySharma-bykr
Автор

Sir, how do I know which row and which column refers to dog, cat or fox? I mean how to know row 1, 2 or 3 refers to a dog?

sadabratakonar
Автор

What is correct confusion matrix, now I am confused, In some places FP become FN & vice versa

MyRohit
Автор

in the report
suppose:

pre recall f1-score support
a
b
c 0.0 0.0 0.0 3


what do mean by the above report
it has an accuracy of more than 0.6

remaining two classes predicted the precision and recall score with support

ravikumarpawar
Автор

How do I plot the confusion matrix for multiclass image classification(say 4 class) while using ImageDataGenerator for augmentation in CNN algorithm?

emyeethiopia
Автор

Hello Sir,
I am doing coding for the breast cancer dataset with Keras Sequential Model.
while printing classification_report and confusion_matrix it says the following error
''Classification metrics can't handle a mix of binary and continuous targets'''. Previously, I was able to do the same, but now showing this error. Kindly help me out.

SandeepSingh-xery
Автор

For multiclass classification which parameter will be more important precision or recall and why??

ankitashukla
Автор

Sir, please make a video on ROC and AUC with practical example..!!

ParamChoksi-veyg
Автор

Hi what is support in the classification metrics report and what its equation also, what are micro average and other values in the below of report.

aseelsami.
Автор

sir you have explained Accuracy and precision pretty well but Recall and other performance matrix are not explained only cursory introduction is given.

vijai
Автор

Krish, please upload further Deep Learning videos.

sachinborgave
Автор

Is it possible for precision, recall and f1-score to be zero? if yes then what does that mean

siphendulwezaza
Автор

simple question:
You said when we have imbalanced data set, we cannot always go for accuracy, we have to consider precision or recall to judge the accuracy. Now the above final matrix which we have how can we find accuracy here, is it the F1 score we should consider, but there are 3 classes. I am a little confused here

aashishraina
Автор

Dear sir, I am using Rbf. I want to know how to select the hyper parameters. please kindly help

jayasimhayenumaladoddi
Автор

Sir plz start classes on 10th standard Artificial intelligence plz

gamingterrorov
Автор

Sir can you please upload video on indepth intuition of logistic regression cost function. It’ll really help to understand concept in better way

sagarmestry