def load_titanic():
load(
dataset_url=
'https://web.stanford.edu/class/archive/cs/cs109/cs109.1166/stuff/titanic.csv',
out_train_dataset_path='data/titanic_train.csv',
out_val_dataset_path='data/titanic_val.csv')
def load_body_fat():
load(dataset_url=
'https://course1.winona.edu/bdeppa/Stat%20425/Data/bodyfat.csv',
out_train_dataset_path='data/bodyfat_train.csv',
out_val_dataset_path='data/bodyfat_val.csv')
def load_boston_housing():
load(dataset_url=
'http://course1.winona.edu/bdeppa/Stat%20425/Data/Boston_Housing.csv',
out_train_dataset_path='data/boston_train.csv',
out_val_dataset_path='data/boston_val.csv')
def run_tabular_regression(client,
csv_file_url,
gpus,
hours,
features=None,
target=None,
queries=None):
'''
Runs a sample full train-inference flow for the task ``TABULAR_REGRESSION``.
'''
task = 'TABULAR_REGRESSION'
# Randomly generate app & model names to avoid naming conflicts
app_id = gen_id()
app = 'tabular_regression_app_{}'.format(app_id)
xgb_model_name = 'XgbReg_{}'.format(app_id)
train_dataset_path = 'data/{}_train.csv'.format(app)
val_dataset_path = 'data/{}_val.csv'.format(app)
print('Preprocessing dataset...')
load(csv_file_url, train_dataset_path, val_dataset_path)
print('Creating & uploading datasets onto SINGA-Auto...')
train_dataset = client.create_dataset('{}_train'.format(app), task,
train_dataset_path)
pprint(train_dataset)
val_dataset = client.create_dataset('{}_val'.format(app), task,
val_dataset_path)
pprint(val_dataset)
print('Adding models "{}" to SINGA-Auto...'.format(xgb_model_name))
xgb_model = client.create_model(xgb_model_name, task, 'examples/models/tabular_regression/XgbReg.py', \
'XgbReg', dependencies={ ModelDependency.XGBOOST: '0.90' })
pprint(xgb_model)
print('Creating train job for app "{}" on SINGA-Auto...'.format(app))
budget = {BudgetOption.TIME_HOURS: hours, BudgetOption.GPU_COUNT: gpus}
train_job = client.create_train_job(app,
task,
train_dataset['id'],
val_dataset['id'],
budget,
models=[xgb_model['id']],
train_args={
'features': features,
'target': target
})
pprint(train_job)
print('Waiting for train job to complete...')
print('This might take a few minutes')
wait_until_train_job_has_stopped(client, app)
print('Train job has been stopped')
print(
'Listing best trials of latest train job for app "{}"...'.format(app))
pprint(client.get_best_trials_of_train_job(app))
print('Creating inference job for app "{}" on SINGA-Auto...'.format(app))
pprint(client.create_inference_job(app))
predictor_host = get_predictor_host(client, app)
if not predictor_host:
raise Exception('Inference job has errored or stopped')
print('Inference job is running!')
if queries is not None:
print('Making predictions for queries:')
print(queries)
predictions = make_predictions(client, predictor_host, queries)
print('Predictions are:')
print(predictions)
print('Stopping inference job...')
pprint(client.stop_inference_job(app))
def load_diabetes():
load(
dataset_url=
'https://raw.githubusercontent.com/plotly/datasets/master/diabetes.csv',
out_train_dataset_path='data/diabetes_train.csv',
out_val_dataset_path='data/diabetes_val.csv')
def load_heart():
load(dataset_url='data/heart.csv',
out_train_dataset_path='data/heart_train.csv',
out_val_dataset_path='data/heart_val.csv')