Tutorial 36- Logistic Regression Mutliclass Classification(OneVsRest)- Part 3| Data Science

[@ankanbhattacharyya8805]

This is pretty helpful. So complecated concept of multiclass classification explained so simply. Thank you so much. Keep it up

[@ahmedaman6384]

Fantastic as always krish. You are by far the best resource for learning data science concepts. Appreciate you.

[@egorgavriushkin9230]

You know thye material so well and you are so talented as a teacher! thank you for your videos. The IBM data science P.C. is no match to you

[@datatorture3086]

Thanks for your best intuition Sir ,This tutorial has weed out all my confusion

[@sandipansarkar9211]

Thanks Krish.I am still wondering why I didn't see this video earlier .Things would have been far more easier.Now got to study some real easy language articles or blogs especially medium / towards data science to have a better idea about this,.

[@thepresistence5935]

cleared everything about the logistic regression so thanks dude

[@hrushikeshkulkarni7353]

Thank you Krish. This is actually very simple to understand 😀

[@cosmin2889]

Thanks man you are pure gold :D

[@kv11gaming]

Please publish a video for the multinomial classification using sklearn for all classification models properly in python, so that we can understand properly.

[@maheshchoudhury2719]

Your video helps me lot... Thanks a lot..

[@RamKrishna-dj5sj]

very nicely explained

[@trivendratiwari9231]

Thank you krish for these videos 👍

[@soumyadeeproy6611]

Hi,I have a small doubt which is: U said for each category, one model is trained which outputs its probability of being correctly classified. Now, when each of the models is independent and thus yields independent probability values, then how you are saying that their sum would be equal to 1 ?? I hope I made my question clear

[@trashantrathore4995]

Did u get any explanation for it from anywhere? If yes, plz let me know

[@r0cketRacoon]

@@trashantrathore4995 through softmax function

[@fadlyahmad8565]

Thank you for the explanation. Its very helpful!

[@aadarshshekhar6645]

Sir, could you please explain the difference between multiclass classification and clustering algorithm.

[@louerleseigneur4532]

Thanks Krish

[@VarunKumar-pz5si]

Great Explanation

[@borispasynkov1404]

In 6 minutes explained the whole material of 1 hour university lecture

[@yannickpezeu3419]

Thanks, really clear explanation

[@jihedjerbi2148]

Great technique ...

[@PavanKumar-ef1yy]

Thank you

[@krishnamishra8598]

Great technique..loved it😘

[@anandkumar-cy3st]

@krish could you please upload a video about logistic regression implementation

[@kasturikasturi2551]

thank you sir

[@prashantjoshi8847]

clean and precise :-)

[@babaabba9348]

GOOD THANK YOU

[@Martinxddxdxdxdx]

Great technique. is the same as use OneVsRestClassifier from sklearn?

[@fastlearningpoint4721]

Good sir

[@datasciencegyan5145]

can u pls make one video related to coding in multi-class classification using logistic regression

[@MuhammadAhmad-bx2rw]

Great

[@thedataguyfromB]

Owesome

[@GauravSharma-ui4yd]

Nice video krish, why we take -1 instead of 0? Also just one thing to add that we normalizes the output probabilities at the end either using linear normalizer or softmax.

[@dragon_warrior_]

just for explanation purpose I guess

[@darshanpadaliya9894]

what we take for o1, o2, o3 is the value of y not the value of sigmoid function

[@mrunaldusane3905]

Sir, for multiclass classification (having more than 2 categories in dependent variable). Can we use Multinomial Logistic Model from GLM family?

[@siddharthgurav6407]

it is like creating dummy variables right

[@divyashetty3181]

So for every new test data output would be o3??

[@adityay525125]

Sir I have one question, please tell us where to practice problems. on Machine Learning, I am starting to fell that theory only is not going to cut it anymore.

[@amoldeepgupta1400]

@db1 make small projects

[@MaLik-gz9vb]

How to calculate probability of m1,2 and 3. Formula ?

[@deepankzanwar2187]

Sir what if we have 3 class in our target variable and we apply one hot encoding and while doing train test split we select one variable out if 3 as our Y and build a model. Likewise we can build 3 model. Is this way correct or not.

[@GauravSharma-ui4yd]

Absolutely correct but at the end it is also recommend to normalize the predicted probabilities from all three classes using a softmax/linear-normalizer or any other function that satisfies the normalization properties. Also in the video krish may have forgotten to add that we generally normalizes the probs at the end, scikit does so using linear normalizer but softmax can also we used just like we did at the output layer of neural-net in mutliclass problems

[@a.mo7a]

@@GauravSharma-ui4yd Hello dear sir, can you please explain what is probability normalization?

[@samardwivedi6090]

sir how can we plot the visualization graph for AUC and ROC curve for a huge multiclass data set

[@samriddhlakhmani284]

Is there something like that ? please let me know

[@Ruhgtfo]

Great tutorial~

[@rohankavari8612]

how the sum of all prob of all models is equal to 1?....I think in some cases it might be greater than one.....for eg if 2 model gives .5 value and 3rd gives 0.3 ....then the sum is 1.3

[@a.mo7a]

Exactly my question... but I tried it in python and got 1 for all samples... which is kinda odd

[@rohankavari8612]

@@a.mo7a check the documentation....it might have divided it with some number

[@frischidn3869]

Sir, the link is not working

[@mgaf7864]

The code link is not working

[@nileshmandlik9662]

implementation video for this

[@arunbharathapu2944]

Do this multi classification in deep learning

[@GauravSharma-ui4yd]

You can easily do so in neural net by applying sigmoid activation at the output layer. But the probs across the classes are not normalized hence you can add a softmax layer after it. But first adding sigmoid and then softmax doesn't makes sense hence we directly apply softmax at the end. Also in this case as well we are recommended to normalize the probs across the classes.

[@gauravtak9787]

Sir i think this is similar as softmax function i am right or not plzz clear me

[@thallamkarthik7020]

What if 2 of those 3 classes got the same probability values? How the model is gonna classify that new test point ?

[@vasanth3029]

Its generally a best practise to get the probabilities and setting threshold manually and deciding the output

[@ghaythalmahadin4994]

what if probabilities are equal, which class will be chosen?

[@sumitmhaiskar722]

point did you get the answer?

[@sumitmhaiskar722]

@Epoxy To The World no sir

[@sumitmhaiskar722]

@Epoxy To The World I'm 23 years old and how may I know that what path you are following???

[@sumitmhaiskar722]

@Epoxy To The World yes or just go with the flow and try to understand the maths behind each and every algorithm.

[@nit235]

This is what I understood after doing some research on this question : Let assume, we have 3 binary classification models, each one has been trained to output whether the input belongs to (1,2,3 respectively) or not. Assume we have a query or test point as input. We need to return the labels for this input. Since, it is a multiclass problem, the input could have 3 labels. Assume we set our threshold to 0.6 . Now, we pass the input to all these 3 models. Assume we got these probabilities from each of the 3 models as follows : 0.52,0.7,0.8 . Since, we made our threshold to 0.6, the input doesn't belong to the first because 0.52 is less than our threshold 0.6 . We now output the labels for the query as 2 and 3 since their probabilities are greater than the threshold. I'm not sure about my answer but this is what I got after googling it.

[@InfinitelyScrolling]

does logistic regression always fit a line or it can fit curves also?

[@revanthshalon5626]

Logistic regression is a linear model fitting a curve or sigmoid so that the value stays positive and between 0 and 1

[@keshavrjaput1522]

Plz send me the ineron course link form

[@raoashwi]

is it 0 0 1 or -1 -1 1