def fit(cls, space=dflt, max_evals=100, mongo=None):
if mongo:
trials = mongoexp.MongoTrials(mongo['url'], exp_key=mongo['exp'])
else:
trials = None
best = hpo.fmin(
cls.objective,
space,
algo=hpo.tpe.suggest,
max_evals=max_evals,
trials=trials
)
return best
model.compile(loss='categorical_crossentropy',
optimizer=RMSprop(),
metrics=['accuracy'])
model.fit(x_train, y_train, batch_size=128, epochs=10, validation_data=(x_test, y_test))
score, acc = model.evaluate(x_test, y_test, verbose=0)
out = {
'loss': -acc,
'score': score,
'status': STATUS_OK,
'model_params': params,
}
# optionally store a dump of your model here so you can get it from the database later
temp_name = tempfile.gettempdir()+'/'+next(tempfile._get_candidate_names()) + '.h5'
model.save(temp_name)
with open(temp_name, 'rb') as infile:
model_bytes = infile.read()
out['model_serial'] = model_bytes
return out
if __name__ == "__main__":
trials = mongoexp.MongoTrials('mongo://*****:*****@mongodb.host:27017/jobs/jobs', exp_key='mnist_test')
best_run, best_model = optim.minimize(model=create_model,
data=data,
algo=tpe.suggest,
max_evals=10,
trials=trials,
keep_temp=True) # this last bit is important
print("Best performing model chosen hyper-parameters:")
print(best_run)
outf.write("Evalutation of best performing model:\n")
outf.write(str(best_eval) + "\n")
#outf.write("Best performing model chosen hyper-parameters:\n")
#outf.write(str(real_param_values) + "\n")
outf.write("Best performing model chosen hyper-parameters index:\n")
outf.write(str(best_run) + "\n")
outf.write("Summary of best model:\n")
best_model.summary(print_fn=lambda x: outf.write(x + '\n'))
if __name__=="__main__":
#x_train,y_train,x_validate,y_validate,x_test,y_test=data_from_bed()
x_train,y_train,x_validate,y_validate,x_test,y_test=data()
#model = create_model0(x_train,y_train,x_validate,y_validate,x_test,y_test)
print("seed=1234")
trials = mongoexp.MongoTrials('mongo://localhost:12345/foo_db/jobs', exp_key='exp1')
best_run,best_model=optim.minimize(model=create_model,
data=data,
algo=tpe.suggest,
max_evals=100,
trials=trials,
rseed=1234,
keep_temp=True)
#from hyperas.utils import eval_hyperopt_space
#space = get_space()
#real_param_values = eval_hyperopt_space(space, best_run)
best_eval = best_model.evaluate(x_test, y_test)
outf=open('hyperas.results.txt','w')
score = model.evaluate(X_test_aug, y_test)
out = {
'loss': -score[2],
'score': score[0],
'status': STATUS_OK,
'model_params': params,
}
# optionally store a dump of your model here so you can get it from the database later
temp_name = tempfile.gettempdir() + '/' + next(
tempfile._get_candidate_names()) + '.h5'
model.save(temp_name)
with open(temp_name, 'rb') as infile:
model_bytes = infile.read()
out['model_serial'] = model_bytes
return out
if __name__ == "__main__":
trials = mongoexp.MongoTrials('mongo://localhost:27017/jobs/jobs',
exp_key='exp1')
best_run, best_model = optim.minimize(
model=create_model,
data=data,
algo=tpe.suggest,
max_evals=10,
trials=trials,
keep_temp=True) # this last bit is important
print("Best performing model chosen hyper-parameters:")
print(best_run)
'pred_y': pred_y
}, pkl_out)
print("RMSE: {} in {}s".format(*metric_results.values()))
return {
'loss': metric_results['rmse'],
'status': STATUS_OK,
'model': params,
'metrics': metric_results,
'figure': Binary(bytes_out.getvalue()),
'pred_data': Binary(pkl_out.getvalue())
}
if __name__ == '__main__':
import sys
print(sys.argv)
trials = mongoexp.MongoTrials(sys.argv[1], exp_key='htm_sm_115_proper')
best_run, best_model = optim.minimize(model=create_model,
data=data,
algo=tpe.suggest,
max_evals=200,
keep_temp=True,
trials=trials)
print("Best performing model chosen hyper-parameters:")
print(best_run)
print(json_util.dumps(best_model, indent=4))
'status': STATUS_OK,
'model': model._updated_config(),
'metrics': {
'rmse': rmse_result,
# 'geh': geh(unscaled_pred, unscaled_test_y)[0],
'duration': (finish - start).total_seconds()
},
'figure': fig_b64,
'params': params
}
if __name__ == '__main__':
# trials = Trials()
import sys
trials = mongoexp.MongoTrials(sys.argv[1], exp_key='lstm_3_layer_115_2')
best_run, best_model = optim.minimize(model=create_model,
data=data,
algo=tpe.suggest,
max_evals=300,
trials=trials)
print(
"BEST_RUN",
json.dumps(
{
k: v
for k, v in trials.best_trial['result'].items()
if k not in ['model', 'figure']
},
indent=4))