Unknown Regularizer: L2 In Tensorflowjs
I have trained a model in python using model reg = 0.000001 model = Sequential() model.add(Dense(24, activation='tanh', name='input_dense', input_shape=input_shape)) model.add(GRU(
Solution 1:
Option 1
There are no classes L1
and L2
; they are just interfaces(more here)
There is a class L1L2
which will take the config and return the right regularizer. You can manually replace all occurences of L2
to L1L2
.
Option 2
Register a class L2
classL2 {
static className = 'L2';
constructor(config) {
return tf.regularizers.l1l2(config)
}
}
tf.serialization.registerClass(L2);
// now load the model
Post a Comment for "Unknown Regularizer: L2 In Tensorflowjs"