KerasClassifier object has no attribute model

Question

I'm using CalibratedClassifierCV to calibrate the probabilities of my CNN model. I'm using the following code:

from tensorflow.keras.models import load_model
from tensorflow.keras.wrappers.scikit_learn import KerasClassifier
from sklearn.calibration import CalibratedClassifierCV
def load_model(*args, **kwargs):
    path="my_model.hd5"
    model = load_model(path)
    return model

clf = KerasClassifier(build_fn=load_model)
calib = CalibratedClassifierCV(clf, cv='prefit', method='sigmoid')
calib.fit(X_train, y_train)

When using this code I'm getting the error message AttributeError: 'KerasClassifier' object has no attribute 'model'. Also when I use clf.predict(X_test) I'm getting the same error. So something seems to be wrong with KerasClassifier.

Is there a mistake in my code?

David M. · Accepted Answer · 2021-01-05 20:44:57Z

1

You redefined keras' load_model function (from tensorflow.keras.models import load_model) with a namesake (def load_model(*args, **kwargs)) - that may be the problem.

answered Jan 5, 2021 at 20:44

David M.

4,6383 gold badges22 silver badges28 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

KerasClassifier object has no attribute model

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related