"mlflow.user": "******"
}
run.set_tags(tags)
job_info_dict = {
"run_id": run._run_id,
"experiment_name": exp.name,
"experiment_id": exp._id
}
json_dict = json.dumps(job_info_dict)
with open(job_info_path, "w") as f:
f.write(json_dict)
f.close()
# log environment variables
env_dictionary["MLFLOW_EXPERIMENT_ID"] = exp._id
env_dictionary["MLFLOW_RUN_ID"] = run._run_id
env_dictionary["MLFLOW_TRACKING_URI"] = _get_mlflow_tracking_uri(ws)
env_dictionary["HOME"] = "~/"
print("Before running train")
try:
print("Trying to run train file ")
ret, _ = run_command([sys.executable] + sys.argv[3:],
env=env_dictionary)
except subprocess.CalledProcessError as e:
print("Subprocess caused error " + run_name)
better_e = RuntimeError("{}\n{}".format(e.stderr, e))
run.fail(error_details=better_e)
raise better_e
else:
run.complete()
print("Marked as complete")
# モデルの評価
import os, json
from azureml.core import Workspace
from azureml.core import Experiment
from azureml.core.model import Model
import azureml.core
from azureml.core import Run
run_id = "" # failするrun id を記載
# Workspaceの取得
ws = Workspace.from_config()
# 最新のrun, Experimentの取得
with open("aml_config/run_id.json") as f:
config = json.load(f)
new_model_run_id = config["run_id"]
experiment_name = config["experiment_name"]
exp = Experiment(workspace=ws, name=experiment_name)
fail_run = Run(exp, run_id=run_id)
fail_run.fail()