def test_trasformer_pytorch_logger(integration_test_url, project_name,
use_google_oauth):
merlin.set_url(integration_test_url, use_google_oauth=use_google_oauth)
merlin.set_project(project_name)
merlin.set_model("transformer-logger", ModelType.PYTORCH)
model_dir = "test/transformer"
undeploy_all_version()
resource_request = ResourceRequest(1, 1, "100m", "200Mi")
transformer = Transformer(
"gcr.io/kubeflow-ci/kfserving/image-transformer:latest",
resource_request=resource_request,
)
logger = Logger(
model=LoggerConfig(enabled=True, mode=LoggerMode.ALL),
transformer=LoggerConfig(enabled=True, mode=LoggerMode.ALL),
)
with merlin.new_model_version() as v:
merlin.log_pytorch_model(model_dir=model_dir)
endpoint = merlin.deploy(transformer=transformer, logger=logger)
assert endpoint.logger is not None
model_config = endpoint.logger.model
assert model_config is not None
assert model_config.enabled
assert model_config.mode == LoggerMode.ALL
transformer_config = endpoint.logger.transformer
assert transformer_config is not None
assert transformer_config.enabled
assert transformer_config.mode == LoggerMode.ALL
with open(os.path.join("test/transformer", "input.json"), "r") as f:
req = json.load(f)
sleep(5)
resp = requests.post(f"{endpoint.url}", json=req)
assert resp.status_code == 200
assert resp.json() is not None
assert len(resp.json()["predictions"]) == len(req["instances"])
model_endpoint = merlin.serve_traffic({endpoint: 100})
sleep(5)
resp = requests.post(f"{model_endpoint.url}", json=req)
assert resp.status_code == 200
assert resp.json() is not None
assert len(resp.json()["predictions"]) == len(req["instances"])
# Try to undeploy serving model version. It must be fail
with pytest.raises(Exception):
assert merlin.undeploy(v)
# Undeploy other running model version endpoints
undeploy_all_version()
def test_custom_model_with_artifact(integration_test_url, project_name,
use_google_oauth):
merlin.set_url(integration_test_url, use_google_oauth=use_google_oauth)
merlin.set_project(project_name)
merlin.set_model("custom-w-artifact", ModelType.CUSTOM)
undeploy_all_version()
resource_request = ResourceRequest(1, 1, "1", "1Gi")
model_dir = "test/custom-model"
BST_FILE = "model.bst"
iris = load_iris()
y = iris["target"]
X = iris["data"]
dtrain = xgb.DMatrix(X, label=y)
param = {
"max_depth": 6,
"eta": 0.1,
"silent": 1,
"nthread": 4,
"num_class": 10,
"objective": "multi:softmax",
}
xgb_model = xgb.train(params=param, dtrain=dtrain)
model_file = os.path.join((model_dir), BST_FILE)
xgb_model.save_model(model_file)
with merlin.new_model_version() as v:
v.log_custom_model(
image="ghcr.io/tiopramayudi/custom-predictor-go:v0.2",
model_dir=model_dir)
endpoint = merlin.deploy(v,
resource_request=resource_request,
env_vars={"MODEL_FILE_NAME": BST_FILE})
sleep(5)
resp = requests.post(f"{endpoint.url}", json=request_json)
assert resp.status_code == 200
assert resp.json() is not None
assert resp.json()["predictions"] is not None
model_endpoint = merlin.serve_traffic({endpoint: 100})
sleep(5)
resp = requests.post(f"{model_endpoint.url}", json=request_json)
assert resp.status_code == 200
assert resp.json() is not None
assert resp.json()["predictions"] is not None
# Try to undeploy serving model version. It must be fail
with pytest.raises(Exception):
assert merlin.undeploy(v)
# Undeploy other running model version endpoints
undeploy_all_version()
def test_stop_serving_traffic(integration_test_url, project_name,
use_google_oauth):
merlin.set_url(integration_test_url, use_google_oauth=use_google_oauth)
merlin.set_project(project_name)
merlin.set_model("stop-serving-traffic", ModelType.SKLEARN)
model_dir = "test/sklearn-model"
MODEL_FILE = "model.joblib"
undeploy_all_version()
with merlin.new_model_version() as v:
clf = svm.SVC(gamma='scale')
iris = load_iris()
X, y = iris.data, iris.target
clf.fit(X, y)
dump(clf, os.path.join(model_dir, MODEL_FILE))
# Upload the serialized model to MLP
merlin.log_model(model_dir=model_dir)
endpoint = merlin.deploy(v)
sleep(5)
resp = requests.post(f"{endpoint.url}", json=request_json)
assert resp.status_code == 200
assert resp.json() is not None
assert len(resp.json()['predictions']) == len(request_json['instances'])
model_endpoint = merlin.serve_traffic({endpoint: 100})
sleep(5)
resp = requests.post(f"{model_endpoint.url}", json=request_json)
assert resp.status_code == 200
assert resp.json() is not None
assert len(resp.json()['predictions']) == len(request_json['instances'])
merlin.stop_serving_traffic(model_endpoint.environment_name)
endpoints = merlin.list_model_endpoints()
for endpoint in endpoints:
if endpoint.environment_name == model_endpoint.environment_name:
assert endpoint.status == Status.TERMINATED
# Undeploy other running model version endpoints
undeploy_all_version()
def test_serve_traffic(integration_test_url, project_name, use_google_oauth):
merlin.set_url(integration_test_url, use_google_oauth=use_google_oauth)
merlin.set_project(project_name)
merlin.set_model("serve-traffic-sample", ModelType.SKLEARN)
model_dir = "test/sklearn-model"
MODEL_FILE = "model.joblib"
undeploy_all_version()
with merlin.new_model_version() as v:
clf = svm.SVC(gamma='scale')
iris = load_iris()
X, y = iris.data, iris.target
clf.fit(X, y)
dump(clf, os.path.join(model_dir, MODEL_FILE))
# Upload the serialized model to MLP
merlin.log_model(model_dir=model_dir)
endpoint = merlin.deploy(v)
sleep(5)
resp = requests.post(f"{endpoint.url}", json=request_json)
assert resp.status_code == 200
assert resp.json() is not None
assert len(resp.json()['predictions']) == len(request_json['instances'])
model_endpoint = merlin.serve_traffic({endpoint: 100})
sleep(5)
resp = requests.post(f"{model_endpoint.url}", json=request_json)
assert resp.status_code == 200
assert resp.json() is not None
assert len(resp.json()['predictions']) == len(request_json['instances'])
# Try to undeploy serving model version. It must be fail
with pytest.raises(Exception):
assert merlin.undeploy(v)
# Undeploy other running model version endpoints
undeploy_all_version()
def test_custom_model_without_artifact(integration_test_url, project_name,
use_google_oauth):
merlin.set_url(integration_test_url, use_google_oauth=use_google_oauth)
merlin.set_project(project_name)
merlin.set_model("custom-wo-artifact", ModelType.CUSTOM)
undeploy_all_version()
resource_request = ResourceRequest(1, 1, "1", "1Gi")
with merlin.new_model_version() as v:
v.log_custom_model(image="ghcr.io/tiopramayudi/custom-predictor:v0.2")
endpoint = merlin.deploy(v, resource_request=resource_request)
with open(os.path.join("test/custom-model", "input.json"), "r") as f:
req = json.load(f)
sleep(5)
resp = requests.post(f"{endpoint.url}", json=req)
assert resp.status_code == 200
assert resp.json() is not None
assert resp.json()["predictions"] is not None
model_endpoint = merlin.serve_traffic({endpoint: 100})
sleep(5)
resp = requests.post(f"{model_endpoint.url}", json=req)
assert resp.status_code == 200
assert resp.json() is not None
assert resp.json()["predictions"] is not None
# Try to undeploy serving model version. It must be fail
with pytest.raises(Exception):
assert merlin.undeploy(v)
# Undeploy other running model version endpoints
undeploy_all_version()