Відео

Great video! Sir, your PC or Laptop specification?

Thank you sir really helpful

thanks saptarsi, this video hepled me to understand the SMOTE much better. thanks again

Very good explanation. You deserve more subscription sir. UA-cam shorts might bring some users. Unfortunately, the algorithm works that way. All the best.

Badaa hi khatarnaak subject hain !! 🙂🙂

Wring at 10.40 s1 and s2 matrix both calculation wrong .

Sir any idea VGG on noise samples?

Thanks for your video. could you please help me to find the next video of that which you discussed about different methods of smote?

Sir for applying ANOVA for feature selection we need to apply normality test to demonstrate whether our data follows a normal distribution? Or we can apply for any type of data set without checking the normal distribution.... could you please clarify it?

Wow, this video on Linear Discriminant Analysis (LDA) by Instructor Saptarsi Goswami is incredibly informative and well-explained. I truly appreciate how he breaks down the concepts and compares LDA to PCA, highlighting the advantages of LDA. The explanations of the Fisher discriminant ratio, inter-class scatter, within-class scatter, and eigenvalue decomposition have given me a solid understanding of LDA. Thank you, Instructor Saptarsi, for sharing your expertise and helping me improve my knowledge in this area!

Hey @Saptarsi Amazing!! how to get your contact details.

English too painful to listen I'm afraid.

In SMOTE, on which bases do we identify a point from minority class, in step 1?

Did anyone had a case where SMOTE made ML models performance even worse?

Thanks Sir , It is very useful.

Thank you sir

Hello sir, how can I do it if I want to solve a problem with three categories with the smote algorithm

Excellent exposition, thanks.

Is mahalanobis and Manhatten is same?

great video. I can't understand how scores gets calculated can you help me??

Wow this is one of the most insightful deep explanations on the origins and mechanics of GMM, EM algorithm. Great job.

how can i add random_state to Agglomerative clustering...my clustering numbers keeps changing everytime i re-run the model

Thanbk you very much sir

top tier and simple

In the SWAP stage, do we drop the medoid node that is replaced by non-medoid node so it is not SWAP checked against other medoid nodes? in other words, dose the medoid node becomes non-medoid if it is replaced by a non-medoid and SWAP checked against the other initial medoids or removed from the list of medoids and non-medoids?

Hi. May i know why the no of components for LDA is based on no of classes when we're trying to reduce the no of features?

What ist X_train1 and y_train1? you use it but it was never defined

Nice explanation of anova lectures that you have given

great job...

Thank you sir.

Sir how to use vgg 16 for gray scale images

Thank you for sharing this video. I have learned much from it.

excellent video...just a small correction, the mean of the y componets of X1 is 3.6 and not 3.8

i need the material

It was a very helpful lecture. Thank you, Sir.

Finally my doubt is resolved, keep Going Sir

Can anyone please provide the link of the datatset

why is there 22 in acc = np.empty(22). I mean can we put some lower no instead of 22? i am stuck in re-training labelled and pseudo-labelled data

Thank you for this video sir.

thank you so much for the video but i have a question in adasyn we choose must far instance to have higher chance of being sampled to avoid over fitting ? if yes but what if was a noisy one ?

Thanks very informative but question is that can we use any csv file with tsne?

thank you , very helpful

You don't share the notebook of this code sir

Very informative and clear. Thank you for your effort! The following are the steps for the self-learning algorithm. 1. Train a supervised classifier on the labelled data. 2. Use the resulting classifier to make predictions on the unlabelled data. 3. Add the most confident of these predictions to the labelled data set. 4. Re-train the classifier on both the original labelled data and the newly obtained pseudo-labelled data. 5. Repeat steps 2-4 until no unlabelled data remain. There are two hyperparameters to set, the maximum number of iterations and the number of unlabelled examples to add at each iteration. One issue of self-learning is if we add many examples with incorrect predictions to the labelled data set, the final classifier may be worse than the classifier only trained on the original labelled data. I hope this answer may help someone interested in semi-supervised learning.

Thanks a lot sir. I have a doubt. I used LDA on Water Potability dataset from Kaggle. I did all the cleaning of data and proceeded with the methods you taught. Since my data was of binary classification, I had only 1 component. So, which graph should I plot for such cases to show classification and how? Also, the accuracy for my dataset was quite low around 61%. Any suggestions on this low accuracy on why it is so?

Sir, it's fantastic. Thnx a lot

Thanks Sir, for the video which was very easy to understand. However I was thinking if the labelled dataset contains sample of 2 classes only(does not contain a sample of a possible 3rd class) and the unlabeled sample contains that specific sample of 3rd class (without the class), then I think the classifier trained on the labeled data cannot properly predict and its confidence for both classes would be low. Can anything or strategy be adopted in this case?

Thankyou very much for such a informative video

The new point I learnt is to calculate the cluster purity with respect to ground labels. Thanks for the knowledge sharing and please keep going!

Thank you sir for giving a wonderful lecture. Can you tell me how I can put the sampling ratio as per my choice instead of 1:1 using SMOTE?