Precision, Recall, F1 score, True Positive|Deep Learning Tutorial 19 (Tensorflow2.0, Keras & Python)

[@codebasics]

Check our Deep Learning Course (in PyTorch) with Latest, Industry Relevant Content: tinyurl.com/4p9vmmds Check out our premium machine learning course with 2 Industry projects: codebasics.io/courses/machine-learning-for-data-science-beginners-to-advanced

[@rahulranjan8682]

I have 4 years of experience in data domain and whenever I go for interview I come to this video to revise this concept. This is actually the best video on KZbin available for this topic.

[@harperjmusic]

This is the best explanation on performance metrics that I've found so far.

[@codebasics]

Glad you liked it John

[@aravinthmegnath3569]

Honestly this is the best explanation I have ever seen. I have been studying ML for the past 3 months and have gone through many tutorials. This video straight away cleared my confusions about the confusion matrix. Thank You so much Dev !

[@longTomm]

Finally. An explanation of precision and recall that makes sense! Great stuff

[@mischievousmaster]

This was some seriously brilliant explanation. Takes patience and passsion. Thank you.

[@codebasics]

Glad it was helpful!

[@kishlayamourya3141]

Thanks a lot!!! Only video that clearly explained precison and recall of BOTH the classes. I really gets confused after observing the results of sklearn classification report as mostly I was explained that positives are important den go for precision and if neg imp go for recall ...also tried to understand by formula but no use. Finally its clear. Thanks again.

[@SeekingTruth0007]

Simple and straightforward explanation, Thank you very much, sir. My only suggestion is to put two or more topics in a slide and discuss them together, such as when relating the confusion matrix concept to precision and recall.

[@rndtest8691]

Awesome explanation ..... These are ever confusing topics .... Which are made so simple in this tutorial...

[@classicemmaeasy2292]

You are a real gem for this well detailed explanation

[@bhaskargg6018]

Your explanation is so clear and the complex confusion matrix concept is Clearly understood, Thank you

[@Spielerandom]

Thank you so much! 🙌🏻Very well explained. From Germany.

[@meriemkh4382]

I swear this guy's the best

[@muhammadwaheedkhan5244]

Nice explained. A suggestion -> also discuss the *WHENs* of each concept you explain. Like, when we use Precisio, Recall, and F1-Score. By the way, great work.

[@gopiruby]

Mind blowing explanation bro. Superrrrb.. you nailed the concept. Thank you so much

[@JapiSandhu]

This is the most intuitive video on this topic. Thanks.

[@jyotigautam992]

This is the best explanation anyone could ask for! 😊

[@ahmedadnanaziz2005]

It is one of the best content ever seen 👏

[@bluesky5166]

Good explanation about precision and recall.

[@vivekverma9908]

This is the way to explain all these measures. Nice video

[@greatfeedbackguy]

Very well explained with precision, hence really easy to recall!

[@codebasics]

😀😀👍🏼

[@sandiproy330]

Wonderful explanation. Very easy to understand. Nice clear step-by-step approach. Thanks for providing a such nice tutorial. thank you.

[@codebasics]

Glad it was helpful!

[@adityasharma2667]

Wondering who are these people who unlike this video!! Very Well explained!!! Thank you

[@codebasics]

Glad it was helpful!

[@bigknp6211]

Thank you for such a great tutorial video. As you said on earlier on start video of your deep learning video series, concept are explained in very simple approach. It remind me the quote by A.E where he said "If you cannot explain in simple way , it means you didn't understood"

[@zurraiinrazali5369]

Yup, finally found the best explanation on classification report.. + easy to understanddd

[@abhishekranjan2617]

you are a true genious sir.... Your way of teaching and explanation is awesome sir.. I love this video😎😎😘😘☺☺👌👌❤❤🥰🥰😍😍🙏🙏🙏🙏🙏🙏

[@karlkfoury2213]

why genius lol

[@mathymessy2193]

Thank you am 😊 happy and submitting my assignment and presenting at work tomorrow

[@devarapallivamsi7064]

Excellent explanation, would have been complete if you have also explained why we need F1 score along when we have precision and recall in place. Thank you for the concise and to the point illustration.

[@abdullahficici4298]

Crystal clear explanation. Thank you

[@codebasics]

Glad it was helpful!

[@jassmalik4594]

Honestly it helped me a lot 🐱

[@meysamjavadzadeh]

best confusion matrix explanation video . thanks👍

[@dexterdev]

0:47 2:29 3:23 3:56 4:51 5:38 5:57

[@phuonglethithanh8498]

Thank you for the explanation on precision and recall. Very easy to understand ❤

[@tanvirhasanmonir1627]

Amazing explanation as always. Thanks a lot for consistently providing quality content.

[@shashwattiwari9666]

mast kaam krta h codebasics bhai tum

[@HabiburRahamaniit]

Super Excellent explanation, Dhaval, Sir.

[@uchennanwosu5327]

Excellent exposition! I have seen some of your other videos in business statistics and they were equally accessible.

[@lllcdert826]

thank you very much the only video i managed to figure out the metrics

[@pankajyadav-en9cw]

The way you have explained these topics is fav bez I didn't get so clarity about these topic thank you explaining it means a lot 🙂

[@codebasics]

Glad you liked it pankaj

[@taufiqulhaque4987]

Each important evaluation metric in a nutshell... Thanks a lot

[@codebasics]

Glad it was helpful!

[@ManikandanRaju]

Beautifully explained. Thank you so much sir

[@694manya]

I have another query: For question 2, let's assume 0 stands for not-rainy and 1 stands for rainy days. A business of a large chain ice cream store is using this model to close some branches if it is predicted to rain. A night before, based on model prediction, employees receive an email if they need to show up tomorrow at work or not. The company is already well-known and doesn't need to acquire more customers. Their focus for now is to save costs and not open the store, when there is no customer due to the rain. With this background, which performance metrics will you choose? (10 Points) a. Accuracy b. Precision c. Recall d. F1_score

[@musicmaniac3264]

Sir you are my hero❤❤.

[@codebasics]

Glad it was helpful!

[@HotshotArafath]

Sir can we apply this metrics in to the potato leaf disease detection project? And should we apply them before or after building the cnn model?

[@ramyaabin5548]

Sir you are Awesome. pls join as Amrita School of Computer Science Engineering we need lecturers like you.

[@nikhilkshaji5165]

Thank for helping to understand these concepts.🥰❤

[@rishavbhattacharjee7182]

Brilliant Deep Learning Series sir😍

[@gauravpathak560]

Thanku sir It's help me to understand the behind the scene related to f1_score,precision and recall. Thanku so much again...

[@Susa270]

I am so Thankful to you. Crisp and Clear. 🙏 🙂

[@codebasics]

You are so welcome

[@jaysoni7812]

I understand these things, but I have question that when and why we use these things?

[@devinpython5555]

in classification task this concepts are useful ,say you have to evaluate model and report its measure you can use this

[@SR-ng5kw]

For more business use cases check this video: kzbin.info/www/bejne/eJynZ4iaoKedq5o

[@LifeAsWeSee678]

Thank you very much for explaining. I just couldn't get what does support means in the end of the video like at 11:07 . Thank you

[@leilaamini405]

awesome. thanks a lot, you are very professional and teach very clear thank you

[@vanisameerayerramilli5049]

Brilliant Explanation!! Thanks a lot :)

[@rafibasha4145]

@2:16 ,it should be False Negative which is predicting dog as not dog .@3:22 it should be False Positive which is predicting No dog as dog here its 3 Please check and correct it

[@adityawanjari5128]

Thank you for such wonderful explanation

[@kalimuthudurairaj9191]

Good content... Nicely explained... Thank you...

[@codebasics]

I am happy this was helpful to you.

[@sushantpenshanwar8038]

Positive is one class (Dog) and negative is another class. Now its easier to remember if we just say in our head True dog vs False Dog instead of True positive and False Positive. Vice versa for other class. Replacing positive and negative with the respective class name.

[@abdelhafiddahhani2514]

Great explanation 😄

[@sanahnahk7312]

Thank you very much for making it simple.

[@gayathrisarangika2512]

Thankyou very much it is very clear explanation ❤❤

[@ayoolafakoya9841]

You are an awesome teacher

[@avani5381]

Wonderful explanation sir, kindly do some example in R language, that was so helpful for our work sir. Your way of explaination is extraordinary sir, support us with R language also . Once again Thank you for the picture example for our understanding.

[@PriyankaDarshanam]

if you get an error saying 'sns' not defined, add "import seaborn as sns" to you code

[@currentactivities6527]

want to thank you, sir, this helps a lot, love from Pakistan.

[@codebasics]

😊👍

[@developerboy8341]

Absolutely great presentation.

[@codebasics]

Glad it was helpful!

[@kekkettoful]

Hello great video, apply the following classifiers: KNN, K-Me, Bayes, Binary Neural Networks and make a comparison between the results with the various techniques (through precision, recall, loss), also showing the confusion matrices, so as to report with which technique I get the best results. can you give me some advice please? Dataset is the 20 newsgroup text

[@aniketmishra2742]

Best explanation ❤

[@sjelallen]

Great explanation! Very helpful!

[@leamon9024]

Always thanks for your effort and hard work.

[@codebasics]

My pleasure!

[@chetanguduru]

Great expaination. Thanks for this!

[@MrAconfee]

Thank you! Great explanation.

[@ilmu011]

One more question, in this example what would the "mean" average precision (mAP) be? And what's the difference between mAP and macro avg and micro avg

[@kunchammadhukumar4820]

damn...! your explanation 🔥🔥🔥

[@ankitachaudhari99]

Subscribed. Explained well!!

[@codebasics]

Glad it was helpful!

[@azamshakori6926]

It was amazing, Thank u so much

[@prudhvivalintine]

Great Visual representation.

[@codebasics]

I am happy this was helpful to you.

[@muskankuchhal2403]

Thank you so much for this amazing content. Sir, can you make the CNN tutorial? If yes, when can we expect it?

[@codebasics]

Yes, soon

[@kashifiqbal5151]

@@codebasics i am also waiting sir..

[@DataLift]

on 3:03 *True Negative = 2* and *False Negative = 1*

[@Gaunayan209]

This is nice session 👍

[@borngenius4810]

thanks for these excellent videos. My question : do we used these Precsion, Recall , F1 parameters while using cross validaton method to chose a model?

[@jerichokayson2899]

you all probably dont give a damn but does anybody know a tool to get back into an Instagram account?? I was dumb lost my login password. I appreciate any assistance you can offer me.

[@sonnybrecken3818]

@Jericho Kayson instablaster =)

[@jerichokayson2899]

@Sonny Brecken thanks so much for your reply. I got to the site on google and Im waiting for the hacking stuff atm. Looks like it's gonna take quite some time so I will get back to you later when my account password hopefully is recovered.

[@jerichokayson2899]

@Sonny Brecken It worked and I actually got access to my account again. I'm so happy:D Thanks so much you really help me out !

[@sonnybrecken3818]

@Jericho Kayson No problem xD

[@manikandang7864]

Great explanation Sir, Really helpful

[@codebasics]

Glad to hear that

[@mriduldongsare1555]

Very well explained

[@mnal.abdulrahman]

Thank you soooooooo soooo much 😍😍🙏🙏🙏👌you helped me a lot

[@codebasics]

You're welcome 😊

[@ilmu011]

Very good explanation!

[@codebasics]

Glad it was helpful!

[@adamtki]

Excellent video!

[@anmol_seth_xx]

Sir, Can you please tell me from these F1 score, what we can compute about our model, and what to do when we get these values?

[@sadi-90]

What is support,Marco avg and weight avg please explain them

[@kunalsinha962]

Best Explanation. Thankyou Sir

[@codebasics]

Glad it was helpful!

[@mujamilkhan714]

Extremely waiting series🙌🏻

[@codebasics]

Glad it was helpful!

[@uwebosch9460]

I loved it. Thank you

[@Lena-of7wd]

You've explained this so clearly, thank you!!

[@codebasics]

Glad it was helpful!

[@shibbirahmed7929]

Can you plz tell me how I can plot ROC curve for multilabel classification (5 labels of data) to compare various ML models accuracy?

[@utn2609]

Thanks so much! Well explained!

[@codebasics]

Glad it was helpful!

[@elbrenantonio81528]

Can you create classification report for image detection or identification?

[@travelwithus3732]

hi, Thanks for the video, what is meant by macro avg and weighted avg in the classification report?

[@PSTWFB]

#confusion#matrix#machinelearning#deep#precision#recall F1 #score#accuracy#true#positive #negative! kzbin.info/www/bejne/j52pmKaXrcaajpI

[@694manya]

Great explantion. I have one question,"which performance metric should I choose (accuracy, precision, recall and f1_score) if I have no information what the dataset is about?"

[@codebasics]

Do you know if dataset is imbalanced? If yes and if you are not downsampling or upsampling or applying any other techniques to handle imbalance then you need to use f1_score. Using accuracy will give you misleading information

[@694manya]

@@codebasics I'll share the dataset here and the exact question.

[@694manya]

@@codebasics 6.Below you see two columns coming from a classification problem. The first column is the predicted value, the second is the real value. A value of 1 stands for positive/yes, and 0 means negative or no. Find TN, TP, FP, FN. Predicted Values Real Values 0 0 1 1 0 0 0 0 0 0 1 0 1 0 1 1 0 0 0 0 0 1 1 1 0 0 1 1 0 0 0 1 0 0 1 1 1 1 0 0 1 1 0 0 1 1 1 0 1 1 0 0 0 0 1 1 1 1 1 1 0 0 0 0 1 1 1 1 0 0 0 0 0 0 1 1 0 0 1 1 0 0 1 1 0 0 0 1 0 0 1 1 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 0 0 1 1 1 1 0 0 0 1 1 1 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 1 0 0 1 1 1 0 1 1 0 0 1 1 0 0 0 0 1 0 0 0 1 1 1 0 1 1 0 0 1 1 0 1 1 1 1 0 1 1 0 0 1 1

[@694manya]

@@codebasics 8.For question 6, which performance metrics will you choose if we don’t have any information about what is the dataset about? (10 Points) a. Accuracy b. Precision c. Recall d. F1_score e. We better not to choose any performance metrics, when we don't know what data is about

[@pasankalansooriya3124]

"import seaborn as sns" is missing in the code. Add it before running

[@akbarboghani1]

Good explanation. How accuracy and precision are used in the real-world scenario? Can you share some references for the same

[@waedkhwies5550]

May I ask a question please? What are the required specifications for a deep learning training server? Thank you

[@harshwardhanchougule3039]

very well explained!!!!

[@PSTWFB]

Artificial Intelligence, Machine Learning, Deep Learning, Supervised, Unsupervised kzbin.infowYHXPNTTVBE?feature=share