Hi. Everyone. I'm trying to code hyperparameter optimization by using keras tuner.
At first I made layer, node and activation function optimization code
for i in range(hp.Int('num_layers',2,9)):
model.add(keras.layers.Dense(units = hp.Int('units_'+str(i),32,512,16), activation = hp.Choice('a_'+str(i), values=['softmax','relu','tanh','sigmoid'])))
like this.(Of course I made model first)
And then I want to optimize 'Optimizer' and 'learning_rate'
But I can't make the code that optimizes two parameter at the same time.
model.compile(optimizer = hp.Choice('optimizer',['Adam','sgd','rmsprop']),
loss = keras.losses.MeanSquaredError(reduction="auto"),
metrics = ['mae'])
this is my optimizer optimization code.(learning_rate is not added)
I want to add learning_rate optimization code so optimizer and learning_rate parameter is optimized simultaneously
learning_rate = hp.Choice('learning_rate', values = [1e-1, 1e-2])
model.compile(optimizer = keras.optimizers.Adam(learning_rate=hp_learning_rate),
loss = keras.losses.MeanSquaredError(reduction="auto"),
metrics = ['mae'])
Like this, i can make each single parameter optimization.
Could you help me to make double parameter optimization?
Thank you so much!
Check https://www.pyimagesearch.com/2021/06/07/easy-hyperparameter-tuning-with-keras-tuner-and-tensorflow/
Also check https://sicara.ai/blog/hyperparameter-tuning-keras-tuner
Kindly have look at this video
https://www.youtube.com/watch?v=vvC15l4CY1Q
