105 Evaluating A Classification Model 6 Classification Report | Creating Machine Learning Models

preview_player
Показать описание

Рекомендации по теме
Комментарии
Автор

Great! Now I know that I can output a dictionary out of a confusion matrix.

FelipeCampelo
Автор

super good video, nicely done and explained. Definitively subscribing!

micaelmarquez
Автор

Why is there two precision scores, considering this metric is for positives only ?

FelipeCampelo
Автор

Nice video, thanks for that info man!
Qustion: I have dataset with 4000 cats and 4000 dogs, I labeled them as 0 and 1 and each image I flattened to be a row with the pixel values.
the test contains 1000 + 1000.
The shape of x_train (8000, 4096) y_train(8000, )
I tried to use KNN with many types of 'n_neighbors' and yet I get a score of 0.53-0.58
In the report I have: (k = 50)
pred recall f1 support
0 0.54 0.81 0.65 1000
1 0.63 0.32 0.42 1000
acc 0.57 2000
mac 0.59 0.57 0.54 2000
wgt 0.59 0.57 0.54 2000

Any Idea how to improve the score?

jamesgoldman
Автор

You're a thief, stealing other people work

tarek_maza