def test_train_fn():
boto_session = boto3.session.Session(region_name=config.AWS_REGION)
sagemaker_session = sagemaker.Session(boto_session)
current_folder = utils.get_current_folder(globals())
datasets_folder = Path(current_folder, "../datasets").resolve()
datasets.clear_datasets(datasets_folder)
sagemaker_session.download_data(
path=str(datasets_folder),
bucket=config.S3_BUCKET,
key_prefix=config.DATASETS_S3_PREFIX,
)
sys_args = [
"--model-dir",
str(Path(current_folder, "../models").resolve()), "--schemas",
str(Path(current_folder, "../schemas").resolve()), "--data-train",
str(Path(current_folder,
"../datasets/data_train").resolve()), "--label-train",
str(Path(current_folder,
"../datasets/label_train").resolve()), "--data-test",
str(Path(current_folder, "../datasets/data_test").resolve()),
"--label-test",
str(Path(current_folder, "../datasets/label_test").resolve())
]
args = ep.parse_args(sys_args)
ep.train_fn(args)
model_assets = ep.model_fn(args.model_dir)
data = {
'contact__has_telephone': False,
'credit__amount': 433,
'credit__coapplicant': 1,
'credit__duration': 18,
'credit__guarantor': 0,
'credit__installment_rate': 3,
'credit__purpose': 'car',
'employment__duration': 0,
'employment__permit': 'foreign',
'employment__type': 'professional',
'finance__accounts__checking__balance': 'no_account',
'finance__accounts__savings__balance': 'high',
'finance__credits__other_banks': 0,
'finance__credits__other_stores': 0,
'finance__credits__this_bank': 1,
'finance__other_assets': 'real_estate',
'finance__repayment_history': 'poor',
'personal__age': 50,
'personal__gender': 'male',
'personal__num_dependents': 1,
'personal__relationship_status': 'married',
'residence__duration': 4,
'residence__type': 'rent'
}
output = ep.predict_fn(data, model_assets)
assert isinstance(output, dict)
assert False
from dotenv import load_dotenv
import os
from pathlib import Path
from package import utils
current_folder = utils.get_current_folder(globals())
dotenv_filepath = Path(current_folder, '../../.env').resolve()
if dotenv_filepath.exists():
load_dotenv()
AWS_ACCOUNT_ID = os.environ['AWS_ACCOUNT_ID']
AWS_REGION = os.environ['AWS_REGION']
STACK_NAME = os.environ['EXPLAIN_STACK_NAME']
S3_BUCKET = os.environ['EXPLAIN_S3_BUCKET']
DATASETS_S3_PREFIX = 'datasets'
SCHEMAS_S3_PREFIX = 'schemas'
OUTPUTS_S3_PREFIX = 'outputs'
SAGEMAKER_IAM_ROLE = os.environ['EXPLAIN_SAGEMAKER_IAM_ROLE']
GLUE_DATABASE = os.environ['EXPLAIN_GLUE_DATABASE']
GLUE_WORKFLOW = os.environ['EXPLAIN_GLUE_WORKFLOW']
ECR_REPOSITORY = os.environ['EXPLAIN_ECR_REPOSITORY']
DOCKER_CONFIG = '/home/ec2-user/.docker/config.json'
