[FIXED] Parameters of SGDClassifier in sklearn

Issue

I am kind of new to machine learning and I am running sample code from a book about machine learning, the code I am trying to run is here:

from sklearn.linear_model import SGDClassifier
clf = SGDClassifier(loss='log', random_state=1)

but when I call partial_fit() I get the error below:

The 'loss' parameter of SGDClassifier must be a str among {'hinge', 'modified_huber', 'huber', 'squared_error', 'epsilon_insensitive', 'squared_epsilon_insensitive', 'perceptron', 'squared_hinge', 'log_loss'}. Got 'log' instead.

can anyone help me with this.

Thank you.

Looking at the documentation, 'log' is a valid parameter.

Solution

Please use 'log_loss' for the loss parameter, It should solve your problem.

Answered By - Shabnam Salman