ium_444356/predict.py

import json
import mlflow
import sys

input = sys.argv[1]

logged_model = 'mlruns/1/3630068c31924c05a9a04e70ef35e0b8/artifacts/s444409'
loaded_model = mlflow.pyfunc.load_model(logged_model)


with open(f'{logged_model}/'+input) as f:
    data = json.load(f)

loaded_model.predict(data['inputs'])
update dockerfile, add predict.py, jenkinsfile_predict 2022-05-11 18:51:12 +02:00			`import json`
			`import mlflow`
update jenkinsfile_predict 2022-05-11 19:36:36 +02:00			`import sys`
update dockerfile, add predict.py, jenkinsfile_predict 2022-05-11 18:51:12 +02:00
update predict.py, jenkinsfile_predict 2022-05-11 19:33:00 +02:00			`input = sys.argv[1]`

update predict.py 2022-05-11 19:53:52 +02:00			`logged_model = 'mlruns/1/3630068c31924c05a9a04e70ef35e0b8/artifacts/s444409'`
update predict.py, jenkinsfile_predict 2022-05-11 19:33:00 +02:00			`loaded_model = mlflow.pyfunc.load_model(logged_model)`
update dockerfile, add predict.py, jenkinsfile_predict 2022-05-11 18:51:12 +02:00

update predict.py, jenkinsfile_predict 2022-05-11 19:33:00 +02:00			`with open(f'{logged_model}/'+input) as f:`
update dockerfile, add predict.py, jenkinsfile_predict 2022-05-11 18:51:12 +02:00			`data = json.load(f)`

update predict.py, jenkinsfile_predict 2022-05-11 19:33:00 +02:00			`loaded_model.predict(data['inputs'])`