Open digicom1978 opened 4 years ago
try using features_train[:(len(features_train)/100)]
instead of features_train[:int(len(features_train)/100)]
try using
features_train[:(len(features_train)/100)]
instead offeatures_train[:int(len(features_train)/100)]
Doing so throws this error:
TypeError: slice indices must be integers or None or have an __index__ method
More on it over here: #207
Docs show that in 0.22 the default for gamma changed from 'auto' to 'scale'. If you set gamma to auto, you get the lower accuracy expected by the mini quiz https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html
same trouble here Python 3.7, seems to depend on the choice of gamma I tried different parameter for gamme: 'auto', 'scale', 1, 0.1 - none of it passed all the Quiz
-> gamma = 1, C= 10000 passes Quiz #34, but fails the following tests -> gamma = 'auto', C=10000 passes Quiz#34+ Quiz#35, does not pass Quiz#36 (and yes with respect to 0-indexed array), but passes Quiz#37
I am running python 3.8 and it works as follow clf = svm.SVC(kernel='rbf', gamma = 'auto') this line for quiz #32 then walk through the following quizzes it will be working easily
I am running python 3.8 and it works as follow clf = svm.SVC(kernel='rbf', gamma = 'auto') this line for quiz #32 then walk through the following quizzes it will be working easily
This worked!
Hi, I am getting different accuracy after changing kernel of svc from linear to rbf. With linear kernel, I got the same accuracy with lessons, but after applying rbf kernel, my accuracy is 0.8953356086461889 while the accuracy in lesson is 0.616040955631
Can anyone help me understand why this difference happen? and how I could correct the code?
from sklearn.svm import SVC clf = SVC(kernel='rbf')
// make data set smaller features_train = features_train[:int(len(features_train)/100)] labels_train = labels_train[:int(len(labels_train)/100)]
t0 = time() clf.fit(features_train, labels_train) print("training time:",round(time()-t0, 3), "s")
t0 = time() pred = clf.predict(features_test) print("prediction time:",round(time()-t0, 3), "s") print(sum(pred == 1))
from sklearn.metrics import accuracy_score acc = accuracy_score(pred, labels_test) print(acc)