def create_problem(load_data):
Problem = NaProblem(seed=2019)
Problem.load_data(load_data)
#Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_search_space, num_layers=10)
Problem.hyperparameters(
verbose=0,
batch_size=100,
learning_rate=0.001, #lr search: 0.01, lr post: 0.001
optimizer='adam',
num_epochs=50,
callbacks=dict(EarlyStopping=dict(
monitor='val_r2', mode='max', verbose=0, patience=5)))
Problem.loss('mse')
Problem.metrics(['r2'])
Problem.objective('val_r2__last')
Problem.post_training(num_epochs=1000,
metrics=['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
},
TensorBoard=dict(log_dir='{}'.format(
time.time()), )))
if __name__ == '__main__':
print(Problem)
from pprint import pprint
pprint(Problem.space)
# Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_structure, num_cells=3)
Problem.hyperparameters(
batch_size=64,
learning_rate=0.001,
optimizer='adam',
num_epochs=1,
)
Problem.loss('mse')
Problem.metrics(['r2'])
Problem.objective('val_r2__last')
Problem.post_training(num_epochs=1000,
metrics=['r2'],
model_checkpoint={
'monitor': 'val_r2',
'mode': 'max',
'save_best_only': True,
'verbose': 1
},
early_stopping={
'monitor': 'val_r2',
'mode': 'max',
'verbose': 1,
'patience': 20
})
from deephyper.benchmark import NaProblem
from deephyper.benchmark.nas.linearReg.load_data import load_data
from deephyper.benchmark.nas.linearRegMultiInputsGen.load_data import load_data
from deephyper.search.nas.model.baseline.simple import create_search_space
from deephyper.search.nas.model.preprocessing import minmaxstdscaler
Problem = NaProblem()
Problem.load_data(load_data)
Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_search_space)
Problem.hyperparameters(
batch_size=100,
learning_rate=0.1,
optimizer='adam',
num_epochs=10,
)
Problem.loss('mse')
Problem.metrics(['r2'])
Problem.objective('val_r2')
# Just to print your problem, to test its definition and imports in the current python environment.
if __name__ == '__main__':
print(Problem)
Problem.load_data(load_data)
# Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_search_space, num_layers=10)
Problem.hyperparameters(batch_size=100,
learning_rate=0.1,
optimizer="adam",
num_epochs=20)
Problem.loss(
loss={
"output_0": "mse",
"output_1": "mse"
},
weights={
"output_0": 0.0,
"output_1": 1.0
},
)
Problem.metrics({"output_0": ["r2", "mse"], "output_1": "mse"})
Problem.objective("val_output_0_r2")
# Just to print your problem, to test its definition and imports in the current python environment.
if __name__ == "__main__":
print(Problem)
Problem.hyperparameters(
batch_size=8,
learning_rate=0.01,
optimizer='adam',
num_epochs=200,
callbacks=dict(EarlyStopping=dict(
monitor='r2', # or 'val_acc' ?
mode='max',
verbose=0,
patience=5)))
Problem.loss('mse') # or 'categorical_crossentropy' ?
Problem.metrics(['r2']) # or 'acc' ?
Problem.objective('r2__max') # or 'val_acc__last' ?
Problem.post_training(
repeat=1,
num_epochs=1000,
metrics=["mse", "r2"],
callbacks=dict()
# callbacks=dict(
# ModelCheckpoint={
# 'monitor': 'val_r2',
# 'mode': 'max',
# 'save_best_only': True,
# 'verbose': 1
# },
# EarlyStopping={
# 'monitor': 'val_r2',
from deephyper.benchmark import NaProblem
from deephyper.benchmark.nas.linearRegMultiVar.load_data import load_data
from deephyper.search.nas.model.baseline.simple_deep import create_search_space
from deephyper.search.nas.model.preprocessing import minmaxstdscaler
Problem = NaProblem(seed=2019)
Problem.load_data(load_data)
# Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_search_space)
Problem.hyperparameters(batch_size=100,
learning_rate=0.1,
optimizer="adam",
num_epochs=1)
Problem.loss("mse")
Problem.metrics(["r2"])
Problem.objective("val_r2")
# Just to print your problem, to test its definition and imports in the current python environment.
if __name__ == "__main__":
print(Problem)
from deephyper.benchmark import NaProblem
from deephyper.benchmark.nas.mnist1D.load_data import load_data
from deephyper.search.nas.model.baseline.simple import create_search_space
from deephyper.search.nas.model.preprocessing import minmaxstdscaler
Problem = NaProblem()
Problem.load_data(load_data)
Problem.search_space(create_search_space)
Problem.hyperparameters(
batch_size=100,
learning_rate=0.1,
optimizer='adam',
num_epochs=10,
)
Problem.loss('categorical_crossentropy')
Problem.metrics(['acc'])
Problem.objective('val_acc')
# Just to print your problem, to test its definition and imports in the current python environment.
if __name__ == '__main__':
print(Problem)
# Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_structure)
Problem.hyperparameters(
batch_size=20,
learning_rate=0.01,
optimizer='adam',
num_epochs=1,
)
Problem.loss('categorical_crossentropy')
Problem.metrics(['acc'])
Problem.objective('val_acc__last')
Problem.post_training(num_epochs=1000,
metrics=['acc'],
model_checkpoint={
'monitor': 'val_acc',
'mode': 'max',
'save_best_only': True,
'verbose': 1
},
early_stopping={
'monitor': 'val_acc',
'mode': 'max',
'verbose': 1,
'patience': 20
})
from search_space import create_search_space
from deephyper.search.nas.model.preprocessing import minmaxstdscaler
Problem = NaProblem(seed=2019)
Problem.load_data(load_data)
# Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_search_space, num_layers=5)
Problem.hyperparameters(
batch_size=32,
learning_rate=0.001,
optimizer='adam',
num_epochs=20,
callbacks=dict(EarlyStopping=dict(
monitor='r2', # or 'val_acc' ?
mode='max',
verbose=0,
patience=10)))
Problem.loss('mse') # or 'categorical_crossentropy' ?
Problem.metrics(['r2']) # or 'acc' ?
Problem.objective('val_r2__last') # or 'val_acc__last' ?
# Just to print your problem, to test its definition and imports in the current python environment.
if __name__ == '__main__':
print(Problem)
Problem.metrics(["binary_accuracy"]) # or 'acc' ?
# def myacc_with_pred(info):
# from sklearn.metrics import accuracy_score
# import numpy as np
# y_pred = np.array(info["y_pred"])
# y_pred = (y_pred >= 0.5).astype(np.int32)
# y_true = np.array(info["y_true"], dtype=np.int32)
# acc = (y_true.flatten() == y_pred.flatten()).astype(int).sum() / len(y_true.flatten())
# # acc = accuracy_score(y_true, y_pred)
# return acc
Problem.objective("binary_accuracy__max") # or 'val_acc__last' ?
Problem.post_training(
repeat=1,
num_epochs=3000,
metrics=["binary_accuracy"],
callbacks=dict()
# callbacks=dict(
# ModelCheckpoint={
# 'monitor': 'val_r2',
# 'mode': 'max',
# 'save_best_only': True,
# 'verbose': 1
# },
# EarlyStopping={
# 'monitor': 'val_r2',
Problem.load_data(load_data)
Problem.preprocessing(minmaxstdscaler)
Problem.search_space(create_search_space)
Problem.hyperparameters(
batch_size=8,
learning_rate=0.01,
optimizer="adam",
num_epochs=200,
verbose=0,
callbacks=dict(EarlyStopping=dict(
monitor="r2",
mode="max",
verbose=0,
patience=5 # or 'val_acc' ?
)),
)
Problem.loss("mse") # or 'categorical_crossentropy' ?
Problem.metrics(["r2"]) # or 'acc' ?
Problem.objective("r2__max") # or 'val_acc__last' ?
# Just to print your problem, to test its definition and imports in the current python environment.
if __name__ == "__main__":
print(Problem)