Tutorial 34- Performance Metrics For Classification Problem In Machine Learning- Part1

[@Kickass3131]

I have almost completed my masters in data science and they only mentioned F1-Score, never F-beta score. You are a very good teacher!!!

[@saikirangolla2054]

Because they follow old syllabus

[@ebujak1]

F1 -score where Bets=1 is for equal weighting of precision and recall. Beta will change depending if your problem should favor precision (or recall).

[@shyamkale7230]

Because this is Krish that is Oldish

[@rpratiksha6547]

Where did you pursue your masters?

[@ritikgupta6825]

this field is so vast that one institute cant tech each and everything, you have to continously gain knowledge from wherever you find

[@AbdullaNabeel]

he is a hero, not the hero we deserved but the hero we needed

[@vaibhavvaibhav6118]

I don't know how to put in words but a big big THANK YOU Krish for making such amazing videos. These are the kind of explanations that I have been searching over for months.

[@naveendubey2815]

This is the best explanation of Metrices i have ever gone through...Fantastic Job!!

[@preetsandhu2844]

I was confused after my lecture in college, just knew names of method but now got to know when to apply what. Thank you Krish, you are Awesome Teacher

[@SanjaySharma-f1l]

I only started learning AI / ML few weeks back and this explanation was brilliant and you made it so easy to understand the concept.

[@munishasharma5265]

Fantastic explanation. The best thing about this video is you explained where is real (IT) world we can use, which matrix to use when. probably one of your masterpieces.

[@UmairMehmood]

Best explanation. In the university, I just learned the formulas but never understood them. Very clear explanation!

[@zulfiqarali-zq1rg]

Hello sir!I cannot imagine how you are helping me at this time of my final defense.God bless you sir .salute you sir thx from the neighbor.

[@swarnimapandey7297]

thanks a lot, man!! I really needed this as hell in my one of the ml projects.. and here u uploaded this video..

[@Santhosh_39]

Amazing explanation Krish. Definitely I will watch this video again, before when I'm attending interviews. So many things will be clarified in 20 min. of explanation.

[@mohammadarif8057]

Sir bhagwaan apko sari khusiya de...splendid job for sharing such valuable concepts

[@milindkulkarni4335]

This is the best channel i came across so far.

[@vipindube5439]

Thanks for your consistency with your learning and teaching.

[@ishakawade9100]

ANY 10TH STANDARAD STUDENT HERE?THANKS FOR THIS VIDEO I'M JUST BLESSED, THERE IS NO CHANNEL DOING THIS FOR OUR SYLLABUS THANKS A LOT MAN DIDN'T EXPECT SUCH A GREAT TUTORIAL HERE ON YT

[@skywalker0803r]

I finished watching the entire playlist Thank you a lot

[@drishtisharma6843]

superb! many congratulations!

[@abhinavkale4632]

@@drishtisharma6843 where is the part 2?

[@viralgo4195]

@@abhinavkale4632 check out the complete machine learning playlist. Watch video 41 for next part.

[@amitbudhiraja7498]

What is the playlist name?

[@ReshamJaiswal-l8y]

At first I was really confused with where things were going because, a lot of it works well for me if examples are shared to explain the scenario (which were later included - Spam and Cancerous/Non-Cancerous). I have been referring to a lot of notes from many sources and none of them could give me the kind of clarity this video gave me on what metrics to focus on and when. Thanks a ton for sharing this.

[@aish_waryaaa]

He is saving me in my masters and clearing my concepts.Thank You sooo muchhh Sir.GOD Bless You:)

[@aravindetagi4008]

Krish always explains and makes concept simpler and make us remember for eternity

[@sohamajgaonkar3119]

you are a savior bcz.... out college faculties are hopelesss..

[@raneshmitra8156]

This is one of your master class video as of now....Thanks Kris...

[@sweetyporwal8769]

Wonderful explanation very sorted & clear , I never got clear understanding on performance matrix before this. Great job i must say..!!

[@faustopf-.]

Incredible explanation, I just started my evaluation phase in my churn prediction model and this will be really helpful, thank you!

[@syedyunus4204]

I'm waiting for part 3 of "implementing all the Metrics For Classification Problem in ML by taking improper data set" as you mentioned in the above video here 1:05 please Krish. As I'm preparing for the interviews as a fresher and I'm immensely in need of the job now and I have a year of a gap also, So I'm sitting continuously to quench my thirst for covering ML part as quickly as possible. so please Krish make the video.

[@chaya6859]

Thank you so much for deep driving the each concept related to the Data Science/Machine learning 😊😊

[@shreyasrajendrachim5282]

Thank you Krish Sir,for explaining precision and recall so effectively, previously it was very confusing.

[@anirbansengupta1107]

This is the best explanation for performance metric selection in classification problems. Thank you Krish!!

[@piyushhibare]

Actually All these topics are very tough to understand but krish explained it very well.. Thanks a lot Krish

[@hiteshdamani30]

Most Awaited video!!! Thank you soo much sir..... 👍👍👍

[@bassantgamal8873]

Thank you a lot, this is really very simple and clear. This is the best explanation I have ever seen, you are amazing.

[@r7918]

Most excellent video I have ever watched on this topic. Thanks a lot krish, it helped me a lot!!!

[@Letsbetog]

If youtube provides 10000+ subscribe and like buttons , I would have like and subscribed all of them.. THank you very sir for bringing such an amazing content for free.

[@architnarang3232]

Brilliant comprehensive playlist 👍🏻👍🏻

[@akshaysharma224]

Wonderful explanation. Loved your teaching style. Thanks for this amazing content !

[@abhishekranjan2617]

superb, awesome and great video sir... you are a great teacher and a person too.. I love your videos sir👌👌👍❤❤❤🙏🙏🙏🙏🙏🙏🙏🙏😊😊😘😘😍😍🥰

[@raghuiyer6848]

too good kris. very useful and informative video

[@titaniumstrong956]

Best explanation..Hats off Krish

[@nitheesh340]

Krish Sir, you're really a super hero compared to Hrithik Roshan's Krrish. :D Everytime I watch your video, my desire to meet you in person is increasing day by day. Wonderful video as always. You made the concept super clear. Would love to learn more from you. Thank you.

[@lalitalone706]

I really enjoy your lecture, Krish. Thanks a lot for putting this video.

[@BrianIvanCusuanto]

this is nugget! Krish explained so well like crystal clear. thanks!

[@abhiksarkar3611]

Literally the best explanation

[@himanshubhusanrath212]

Loved the explanation. Thank you for sharing this info with such simple explanations

[@anaghakatthe4162]

Sir, I am a graduate research student and currently working on my thesis. This video was very helpful. I wanted to learn about ROC,AUC and kappa. I have to defend in 2 months. When could we expect part2 and part3 video, Sir? Could you please let me know.

[@kamalmalik9529]

Great Sir, I was badly in search to clear all my concepts, I really learned a lot. Thanks Sir

[@sandipansarkar9211]

Great explanation Krish .But we need to remember the things to apply them.Thanks

[@Nancivlog]

best video on youtube.. Thank you

[@sunilkumarganapathy3758]

Superb Kris.. Egarly waiting for next part

[@gustavoramalho9454]

You're an amazing teacher! Thanks for this video

[@_proton5]

Excellent video Krish. Kudos!

[@apoorvaka4247]

Excellent with clear explanation 👌👏

[@swamzvj5045]

Great video! But, how do we mathematically/ practically find out if we have a balanced/ unbalanced dataset?

[@fysys_56]

Thank you so much for teaching this good. I really understood the topics . You are a great teacher. 💖💖

[@kshitijbiyani3777]

kudos to you. Very well explained.

[@surbhisoni5614]

thanks Sir for so much efforts to teach so as to make our concepts clear

[@vidyasagar7860]

Good classification sir about performance metrics for novice learner

[@samutsari5956]

Thank you sir for explaining further the concepts

[@bishwajeet_b_das]

Superb explaination.. sir

[@sreelakshmi4230]

Thank you Krish for Wonderful explanation

[@ashwinibalaji350]

Amazing explanation. Thank you so much :) You are always the best!

[@sumitnaina]

Thanks buddy. Now I have full clarity about recall, precision and f beta.

[@sulochanasandaruwan763]

Hey, krish. It is a great video; I could learn a lot and clear few aughts I had with performance metrics. Thank you .. ! Good luck...

[@ektamanvar4397]

oh my god no words to comment explained very well thanks a lot........

[@shaneebkottakkal3491]

Amazing,Very Intresting

[@NM-vw6xq]

Great explanation Krish! Thank you!

[@debasishpanda5648]

In a pathological database there are 2000 patient records out of which 600 are flu patient records. A disease predictor application accesses that database and predicts 150 flu patients correctly out of 200 flu patients and 300 non-flu patients. Find the following performance parameters. a) Sensitivity b) Specificity c) Accuracy d) Precision

[@jeffvarghese7657]

Great Krish.👍

[@shakib.polock]

this is one brilliant video! thank you very much.

[@bhavyav3859]

Thank you sir..for the efforts you have taken to explain!!

[@ijeffking]

Excellent explanation. Thank you

[@rinikhanna7510]

Very helpful video. Thanks Krish :)

[@priyasinha2251]

Hats off Krish!

[@ayselceferzade8587]

Absolutely understandable! Thanks a lot!

[@gauravpardeshi6056]

very good video krish...

[@nkorochinaechetam2516]

thanks for your explanation, I took notes !

[@milenafassea5822]

Krish you are amazing!!!!

[@pranjalgupta9427]

Nice😊👏👍

[@lalitchaudhari7470]

very very nice video sir, Great Explanation..thanks a lot

[@sohamajgaonkar3119]

mind bogling awesome

[@vibesnovibes6320]

You my friend are godsend 👍

[@venud1800]

Excellent Summary Krish 👍

[@aminurrahmanashik2786]

A great video. Thanks a lot.

[@bharathjc4700]

Thank a ton for your valuable inputs

[@rupareddy9243]

Thanks a lot krish, waiting for part-2&3 videos.

[@rishipatel2221]

Krish as u explain that which Beta value will be taken into consideration as per the influence of Recall & Precision. So y don't we directly use Precision or Recall formula.

[@arjundev4908]

Hi Krish, In the video @16:16 seems there is small correction FP and FN should be interchanged if i am not wrong.. as 1-0 is is FN and 0-1 is FP.... correct me if i am wrong.

[@chiragthakor4775]

1 is for True and 0 is for false and also consider 1 as positive and 0 as negative, so [1,0] is TrueNeg, [0,1] is FalsePos, [1,1] TruePos and [0,0] is FalseNeg

[@iramarshad700]

Kindly upload 1 vs 1 classification problem as well. I really like your videos. It is very well explained

[@manojtaleka954]

Thank you Krish for this nice video

[@ankitachaudhari99]

Well explained.👏

[@teklenegash6201]

that is a very nice tutorial ever to easily understand these parameters. But sir Did you interchange the fn and fp place on your confusion matrix table. just on minute 13:00. because even in your other tutorials and other video lectures they put fn on the right corner and fp on the left bottom. thank you if I am mistaken.

[@goelnikhils]

Amazing Content

[@Naveennani-fc1im]

Please do k means algorithm with a dataset.i saw one of k means algorithm and that is on range(1,11)

[@sahilsrivastava9895]

Very informative video. Thanks for this.

[@vimalshrivastava6586]

Excellent Explaination👍

[@Cricketpracticevideoarchive]

Recall is column sum is 1 and precision is Row sum 1

[@saurabhtripathi62]

Thanks a lot krish, waiting for part-2&3 videos , i am really thank full for your videos, great work man

[@krishnaik06]

Already uploaded just follow the complete ML playlist

[@ruchikalalit1304]

@@krishnaik06 can't find please share link

[@narenderbansal5741]

Hi krish, Very nice explanation. Just only one question. In your point of view, model performance is more important or model accuracy. from my point of view it is model accuracy because for model performance we can create pickle file and can maintain the model performance. Please correct me if i am wrong.

[@shivamchaudhary2353]

I have a doubt Krish, Precision aur recall interchangable nahi ho jayega apni positive aur negative ki definition ke hisab se?? Like in spam detection, agar hum 1-spam(positive) 0-not spam(negative) kare to hum abhi Precision dekhe gae kyuki 1 ka 0 ho jaye to dikkat nahi but agar 0 ka 1 ho gya to dikkat hai lekin ab agar hum 1-not spam(positive) 0-spam(negative) yaha par agar 0 ka 1 ho jaye koi baat nahi but 1 ka 0 nahi hona chahiye yaha par hum log False negative kam karna chahte hai is liye recall dekhe gae assumption par depend karte hai kya sub??....aur karta bhi to mujhe kaise pta chle ga ki jo mai assume kar ra hu wahi samne wala assume kar ra hai

[@adiflorense1477]

7:55 Is the basic calculation of all these metrics (FPR, FNR, Recall, Precision, F beta, ROC AUC) confusion matrix?

[@kathirkathir4689]

Excellent sir.

[@victoremuchay5397]

Thanks you so much, for the great insight!