a situation where accuracy score can mislead us is when the cost of false positives and false negatives are different. For example, in medical diagnosis, a false negative (saying a patient is healthy when they are actually sick) may be much more costly than a false positive (saying a patient is sick when they are actually healthy). In this case, a model that minimizes false negatives may be more desirable, even if it has a lower overall accuracy.
Thanks for the amazing explaination, just a small correction in sklearn the order of tp, tn, fp, fn is different for binary classification then what you quoted. in sk learn it is: "By definition a confusion matrix C is such that Ci,j is equal to the number of observations known to be in group i and predicted to be in group j. Thus in binary classification, the count of true negatives is C0,0, false negatives is C1,0, true positives is C1,1 and false positives is C0,1."
Thanks for this wonderful playlist sir. Your content is always on point and easy to understand. I just have a doubt about multivariable classification models. Like you showed at 27:23 where we our model has predicted 0 items for Type 1 but it is for Type 0. I couldn't get that part. Can someone please explain?
Always wondered by ur thorough knowledge & way of explaining it. U explains everything very deeply & in a anyone will understand. Thanku so much. No. 1 Campus X.
No bro it is correct . At first I also thought it was wrong but it's not. FP and FN would be interchanged if predictions were in rows and Actual were in column. You can crosscheck both the images.
what if I choose (Heart disease = 0) and ( Not a Heart Disease= 1) ? In this situation my confusion matrix answer will be different from you. so when should we choose 1 or 0 value ?
I think confusion matrix is not correct, actual along the col, predicted along the row from sklearn.metrics import confusion_matrix y_true = [1, 0, 1, 0, 0, 1] y_pred = [0, 0, 1, 1, 0, 1] confusion_matrix(y_true, y_pred) array([[2, 1], [1, 2]], dtype=int64) y_pred = [0, 0, 1, 1, 0, 0] confusion_matrix(y_true, y_pred) array([[2, 1], [2, 1]], dtype=int64) Look at this example... Let me know if i am wrong...
Although you will be classifying the data, but you are using sigmoid function in logistic regression algorithm which gives the probability of happening of an event. This way our model is giving continuous values not discrete values as 0 or 1.And regression models give continuous values. Hope this will help. Correct me if I am wrong.
Nitish Sir is Bhagwaan of Machine Learning realm.
No one comes near you sir . Top class teaching skills.
a situation where accuracy score can mislead us is when the cost of false positives and false negatives are different. For example, in medical diagnosis, a false negative (saying a patient is healthy when they are actually sick) may be much more costly than a false positive (saying a patient is sick when they are actually healthy). In this case, a model that minimizes false negatives may be more desirable, even if it has a lower overall accuracy.
Thanks for the amazing explaination, just a small correction in sklearn the order of tp, tn, fp, fn is different for binary classification then what you quoted.
in sk learn it is:
"By definition a confusion matrix C is such that Ci,j is equal to the number of observations known to be in group i and predicted
to be in group j.
Thus in binary classification, the count of true negatives is C0,0, false negatives is C1,0, true positives is C1,1 and false positives is
C0,1."
no words ♥Sir please complete all 100 days
The way you teach is literally awesome. Thank you so much for this lecture.
#confusion#matrix#machinelearning#deep#precision#recall F1 #score#accuracy#true#positive #negative!
th-cam.com/video/YlFgsaxagX0/w-d-xo.html
Thanks for this wonderful playlist sir. Your content is always on point and easy to understand. I just have a doubt about multivariable classification models. Like you showed at 27:23 where we our model has predicted 0 items for Type 1 but it is for Type 0. I couldn't get that part. Can someone please explain?
Always wondered by ur thorough knowledge & way of explaining it. U explains everything very deeply & in a anyone will understand. Thanku so much. No. 1 Campus X.
Correction in sklear confusion matrix is different
Prediction it is 0, 1 - means 26 is true negative, and 29 is true positive
Kindly correct the confusion matrix 19:10 it looks like you have misplaced the FP & FN.
No bro it is correct . At first I also thought it was wrong but it's not. FP and FN would be interchanged if predictions were in rows and Actual were in column. You can crosscheck both the images.
Well explained!! Thanks much for this tutorial!
Superb Explanation❤🙌
is there a way to know for which 6 people the model did mistake
Great Explanation !@ Thankyou so much Sir..
Love you sir❤
Thank you so much sir🙏🙏🙏
Well explained sir ❤
what if I choose (Heart disease = 0) and ( Not a Heart Disease= 1) ?
In this situation my confusion matrix
answer will be different from you.
so when should we choose 1 or 0 value ?
Thank You Sir.
Well explained....thank you so much.
Bharat Ratna🙌
sir apne t-shirt bhi vahi pehnee hey " Whats the score"
sir u are excellent
awesome
now i understand why name of confusion matrix start with confusion😆😆
I think confusion matrix is not correct, actual along the col, predicted along the row
from sklearn.metrics import confusion_matrix
y_true = [1, 0, 1, 0, 0, 1]
y_pred = [0, 0, 1, 1, 0, 1]
confusion_matrix(y_true, y_pred)
array([[2, 1],
[1, 2]], dtype=int64)
y_pred = [0, 0, 1, 1, 0, 0]
confusion_matrix(y_true, y_pred)
array([[2, 1],
[2, 1]], dtype=int64)
Look at this example...
Let me know if i am wrong...
what if we say 100% accuracy ?
Confusion matrix jaisa name waisa kaam🥲
Amazing
Why LR is called regression, even though it is classification algo?
Although you will be classifying the data, but you are using sigmoid function in logistic regression algorithm which gives the probability of happening of an event. This way our model is giving continuous values not discrete values as 0 or 1.And regression models give continuous values.
Hope this will help.
Correct me if I am wrong.
@@anshikasharma6846correct. For binary classification.
Thank you!!!!!!!!!!!!!!!!!!
done
❤❤👏
terrorist example is awesome
Sir Baal kaale kerlo
Kar kiye. Recent wale videos dekho. Acha lage to like kar dena video
@@campusx-official Ha sir dekha mast lagre ho aap kaale baalo me
bhai jese marjee karoo , bs lecture awesome deliver krnaa
@@campusx-official haha, Thug Life.