def mlflow_run(params, run_name="LOCAL_REGISTRY"):
"""
Function to start a run within a Default experiment
:param params: ters used for the run, such as arguments to RandomForest scikit-learn
:param run_name: label for the name of the run
:return: experiment ID and run ID
"""
with mlflow.start_run(run_name=run_name) as run:
# Get the run and experimentid
run_id = run.info.run_uuid
experiment_id = run.info.experiment_id
# Create our model type instance
sk_learn_rfr = RandomForestRegressor(params)
# Log params and metrics using the MLflow APIs
mlflow.log_params(params)
mlflow.log_metric("metric_1", random())
mlflow.log_metric("metric_2", random() + 1)
mlflow.log_metric("metric_3", random() + 2)
# Set the notes for experiment and the Runs
MlflowClient().set_experiment_tag(
experiment_id, "mlflow.note.content",
"This is experiment for getting started with MLflow ...")
MlflowClient().set_tag(
run_id, "mlflow.note.content",
"This Run is for getting started with MLflow Model Registry ...")
# Log the model at the same time
mlflow.sklearn.log_model(sk_model=sk_learn_rfr,
artifact_path="sklearn-model")
# Create sample message artifact
if not os.path.exists("messages"):
os.makedirs("messages")
with open("messages/message.txt", "w") as f:
f.write(gen_random_text())
mlflow.log_artifacts("messages")
shutil.rmtree('messages')
# Create scatter random plot artifacts file and log artifact
for npoints in range(55, 70, 5):
fig, ex = gen_random_scatter_plots(npoints)
temp_file_name = Utils.get_temporary_directory_path(
"scatter-plot-", ".png")
temp_name = temp_file_name.name
try:
fig.savefig(temp_name)
mlflow.log_artifact(temp_name, "scatter_plots")
finally:
temp_file_name.close() # Delete the temp file
return (run_id, experiment_id)
Draxl, C., B.M. Hodge, A. Clifton, and J. McCaa. 2015. "The Wind Integration National Dataset (WIND) Toolkit." Applied Energy 151: 355366.
Lieberman-Cribbin, W., C. Draxl, and A. Clifton. 2014. Guide to Using the WIND Toolkit Validation Code (Technical Report, NREL/TP-5000-62595). Golden, CO: National Renewable Energy Laboratory.
King, J., A. Clifton, and B.M. Hodge. 2014. Validation of Power Output for the WIND Toolkit (Technical Report, NREL/TP-5D00-61714). Golden, CO: National Renewable Energy Laboratory.
"""
if __name__ == "__main__":
# Use sqlite:///mlruns.db as the local store for tracking and registery
mlflow.set_tracking_uri("sqlite:///mlruns.db")
# Load and print dataset
csv_path = "data/windfarm_data.csv"
# Use column 0 (date) as the index
wind_farm_data = Utils.load_data(csv_path, index_col=0)
Utils.print_pandas_dataset(wind_farm_data)
# Get Validation data
X_train, y_train = Utils.get_training_data(wind_farm_data)
val_x, val_y = Utils.get_validation_data(wind_farm_data)
# Train, fit and register our model
params_list = [{
"n_estimators": 100
}, {
"n_estimators": 200
}, {
"n_estimators": 300
}]
import requests
from cls.utils import Utils
import json
(x_train, y_train), (val_x, val_y) = Utils.load_data()
data = val_x[0].reshape(1, -1)
data_json = json.dumps(data.tolist())
# print(data_json)
headers = {'Content-Type': 'application/json; format=pandas-records'}
request_uri = 'http://127.0.0.1:5000/invocations'
if __name__ == '__main__':
try:
response = requests.post(request_uri, data=data_json, headers=headers)
print(response.content)
print('done!!!')
except Exception as ex:
raise (ex)