Applying Multi-Class Classification instead of binary classification

[alex80]

I working on machine learning classification problem for logistic regression with 2 classes (walking and sitting).
I am trying to change it to Multi-Class Classification with six activities (classes) (walking, walkingupstairs, walkingdownstairs, sitting, standing, lying) instead of binary classification.


the ML model is defined in the following two classes

1 2 3 4 5

def set_weights(intercept, coef, classes, model=linear_model.SGDClassifier()):     model.intercept_ = intercept     model.coef_ = coef     model.classes_ = classes     return model

1 2 3 4 5 6 7 8

def train_model(intercept_init, coef_init, X, y, epochs, lr, batch_size=None, randomise=True):     if batch_size is None or batch_size <= 0:         batch_size = X.shape[0]     classes = np.unique(y)     model = linear_model.SGDClassifier(loss='log', learning_rate='constant', eta0=lr, verbose=0)     set_weights(intercept_init, coef_init, classes, model)     batch_train(model, X, y, classes, epochs, batch_size, randomise)     return model

Please let me know how to make it work for multi-class ?

[alex80]

Any suggestion ???