def test_full_problem(self):
from deephyper.problem import NaProblem
from deephyper.search.nas.model.preprocessing import minmaxstdscaler
pb = NaProblem()
def load_data(prop):
return ([[10]], [1]), ([10], [1])
pb.load_data(load_data, prop=1.0)
pb.preprocessing(minmaxstdscaler)
def search_space(input_shape=(1, ), output_shape=(1, )):
return
pb.search_space(search_space)
pb.hyperparameters(
batch_size=64,
learning_rate=0.001,
optimizer="adam",
num_epochs=10,
loss_metric="mse",
)
with pytest.raises(NaProblemError):
pb.objective("r2")
pb.loss("mse")
pb.metrics(["r2"])
possible_objective = ["loss", "val_loss", "r2", "val_r2"]
for obj in possible_objective:
pb.objective(obj)
# wrong_objective = ["mse", "wrong", "r2__last__max", "val_mse"]
# for obj in wrong_objective:
# with pytest.raises(WrongProblemObjective):
# # print("obj: ", obj)
# pb.objective(obj)
pb.post_training(
num_epochs=2000,
metrics=["mse", "r2"],
callbacks=dict(
ModelCheckpoint={
"monitor": "val_r2",
"mode": "max",
"save_best_only": True,
"verbose": 1,
},
EarlyStopping={
"monitor": "val_r2",
"mode": "max",
"verbose": 1,
"patience": 50,
},
),
)
def test_full_problem(self):
from deephyper.benchmark import NaProblem
from deephyper.search.nas.model.preprocessing import minmaxstdscaler
pb = NaProblem()
def load_data(prop):
return ([[10]], [1]), ([10], [1])
pb.load_data(load_data, prop=1.)
pb.preprocessing(minmaxstdscaler)
def search_space(input_shape=(1,), output_shape=(1,)):
return
pb.search_space(search_space)
pb.hyperparameters(
batch_size=64,
learning_rate=0.001,
optimizer='adam',
num_epochs=10,
loss_metric='mse',
)
with pytest.raises(NaProblemError):
pb.objective('r2')
pb.loss('mse')
pb.metrics(['r2'])
possible_objective = ['loss', 'val_loss', 'r2', 'val_r2']
for obj in possible_objective:
pb.objective(obj)
wrong_objective = ['mse', 'wrong', 'r2__last__max', 'val_mse']
for obj in wrong_objective:
with pytest.raises(WrongProblemObjective):
pb.objective(obj)
pb.post_training(
num_epochs=2000,
metrics=['mse', 'r2'],
callbacks=dict(
ModelCheckpoint={
'monitor': 'val_r2',
'mode': 'max',
'save_best_only': True,
'verbose': 1
},
EarlyStopping={
'monitor': 'val_r2',
'mode': 'max',
'verbose': 1,
'patience': 50
})
)