Feature Importance Chart in neural network using Keras in Python -
i using python(3.6) anaconda (64 bit) spyder (3.1.2). set neural network model using keras (2.0.6) regression problem(one response, 10 variables). wondering how can generate feature importance chart so:
def base_model(): model = sequential() model.add(dense(200, input_dim=10, kernel_initializer='normal', activation='relu')) model.add(dense(1, kernel_initializer='normal')) model.compile(loss='mean_squared_error', optimizer = 'adam') return model clf = kerasregressor(build_fn=base_model, epochs=100, batch_size=5,verbose=0) clf.fit(x_train,y_train)
at moment keras doesn't provide functionality extract feature importance.
you can check previous question: keras: way variable importance?
or related googlegroup: feature importance
spoiler: in googlegroup announced open source project solve issue..
Comments
Post a Comment