def test_yaml_metadata_missing_fields(tmp_path, config_yaml, request, test_case_number):
config_yaml = request.getfixturevalue(config_yaml)
with open(os.path.join(tmp_path, MODEL_CONFIG_FILENAME), mode="w") as f:
f.write(config_yaml)
if test_case_number == 1:
conf = read_model_metadata_yaml(tmp_path)
with pytest.raises(
DrumCommonException, match="Missing keys: \['validation', 'environmentID'\]"
):
validate_config_fields(
conf,
ModelMetadataKeys.CUSTOM_PREDICTOR,
ModelMetadataKeys.VALIDATION,
ModelMetadataKeys.ENVIRONMENT_ID,
)
elif test_case_number == 2:
with pytest.raises(DrumCommonException, match="Missing keys: \['negativeClassLabel'\]"):
read_model_metadata_yaml(tmp_path)
elif test_case_number == 3:
with pytest.raises(
DrumCommonException,
match="Error - for multiclass classification, either the class labels or a class labels file must be provided in model-metadata.yaml file",
):
read_model_metadata_yaml(tmp_path)
elif test_case_number == 4:
with pytest.raises(
DrumCommonException,
match="Error - for multiclass classification, either the class labels or a class labels file should be provided in model-metadata.yaml file, but not both",
):
read_model_metadata_yaml(tmp_path)
elif test_case_number == 100:
read_model_metadata_yaml(tmp_path)
def setup_validation_options(options):
model_config = get_metadata(options)
validate_config_fields(model_config, ModelMetadataKeys.VALIDATION)
if model_config["type"] == "training":
return _setup_training_validation(model_config, options)
elif model_config["type"] == "inference":
return _setup_inference_validation(model_config, options)
else:
raise DrumCommonException("Unsupported type")
def test_yaml_metadata_missing_fields(custom_predictor_metadata_yaml,
tmp_path):
with open(os.path.join(tmp_path, MODEL_CONFIG_FILENAME), mode="w") as f:
f.write(custom_predictor_metadata_yaml)
conf = read_model_metadata_yaml(tmp_path)
with pytest.raises(
DrumCommonException,
match="Missing keys: \['validation', 'environmentID'\]"):
validate_config_fields(
conf,
ModelMetadataKeys.CUSTOM_PREDICTOR,
ModelMetadataKeys.VALIDATION,
ModelMetadataKeys.ENVIRONMENT_ID,
)
def drum_push(options):
model_config = get_metadata(options)
if model_config["type"] == "training":
validate_config_fields(model_config, ModelMetadataKeys.ENVIRONMENT_ID)
_push_training(model_config, options.code_dir)
elif model_config["type"] == "inference":
validate_config_fields(model_config, ModelMetadataKeys.ENVIRONMENT_ID,
ModelMetadataKeys.INFERENCE_MODEL)
validate_config_fields(model_config[ModelMetadataKeys.INFERENCE_MODEL],
"targetName")
_push_inference(model_config, options.code_dir)
else:
raise DrumCommonException("Unsupported type")
