parser.add_argument(
'--model_name', type=str, default='',help='Name you want to give to the model.'
)
parser.add_argument(
'--model_assets_path',type=str, default='outputs',help='Location of trained model.'
)
args,unparsed = parser.parse_known_args()
print('Model assets path is:',args.model_assets_path)
print('Model name is:',args.model_name)
run = Run.get_context()
pipeline_run = Run(run.experiment, run._root_run_id)
pipeline_run.upload_file("outputs/model/model.pth",os.path.join(args.model_assets_path,"model.pth"))
pipeline_run.upload_file("outputs/model/labels.txt",os.path.join(args.model_assets_path,"labels.txt"))
pipeline_run.upload_file("outputs/deployment/score.py","deployment/score.py")
pipeline_run.upload_file("outputs/deployment/myenv.yml","deployment/myenv.yml")
pipeline_run.upload_file("outputs/deployment/deploymentconfig.json","deployment/deploymentconfig.json")
pipeline_run.upload_file("outputs/deployment/inferenceconfig.json","deployment/inferenceconfig.json")
tags = {
"Conference":"Codecamp"
}
model = pipeline_run.register_model(model_name=args.model_name, model_path='outputs/',tags=tags)
print('Model registered: {} \nModel Description: {} \nModel Version: {}'.format(model.name, model.description, model.version))
for n in n_estimators:
run = experiment.start_logging()
run.log("n_estimators", n)
model = preprocessor_pipe.set_params(L_model__n_estimators=n)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
run.log("acc", accuracy)
model_name = "model_n_estimators_" + str(n) + ".pkl"
filename = "outputs/" + model_name
joblib.dump(value=model, filename=filename)
run.upload_file(name=model_name, path_or_stream=filename)
run.complete()
###############
maximum_acc_runid = None
maximum_acc = None
for run in experiment.get_runs():
run_metrics = run.get_metrics()
run_details = run.get_details()
# each logged metric becomes a key in this returned dict
run_acc = run_metrics["acc"]
run_id = run_details["runId"]
if maximum_acc is None:
# COMMAND ----------
model_nm = "bikeshare.mml"
model_output = '/mnt/azml/outputs/' + model_nm
model_dbfs = "/dbfs" + model_output
lrPipelineModel.write().overwrite().save(model_output)
# COMMAND ----------
model_name, model_ext = model_dbfs.split(".")
# COMMAND ----------
model_zip = model_name + ".zip"
shutil.make_archive(model_name, 'zip', model_dbfs)
azRun.upload_file("outputs/" + model_nm, model_zip)
# COMMAND ----------
azRun.register_model(model_name='model_nm', model_path="outputs/" + model_nm)
# COMMAND ----------
# now delete the serialized model from local folder since it is already uploaded to run history
shutil.rmtree(model_dbfs)
os.remove(model_zip)
# COMMAND ----------
mlflow.end_run()
if not (run.id.startswith('OfflineRun')):
# get last section of the mnt location starting after '/workspaceblobstore' which is the blob storage location
blobstoragepath = mntpath.rsplit('/workspaceblobstore', maxsplit=1)[1]
dataset = Dataset.Tabular.from_delimited_files(
path=[(ws.get_default_datastore(), blobstoragepath)])
# getting the model output path (the path to the model.pkl file) from the previous step's output.
model_output = os.environ['AZUREML_DATAREFERENCE_model_output']
print("model path", model_output)
print("files in model path", os.listdir(path=model_output))
# will register the model to the parent which encapsulates all the steps
# if model already exists, error will be thrown, ignore.
try:
# to register a model to a run, the file has to be uploaded to that run first.
parentrun.upload_file('model.pkl', model_output + '/model.pkl')
except:
None
# model_path is the local path of the parentrun model file (see upload_file action above)
model_path = "model.pkl"
print("...working directory")
print(os.listdir(path='.'))
# if the model has been previously registered, retrieve the accuracy. This will be used to determine if the new model
# should be registered or if the old one should be kept as the latest version.
try:
model = Model(ws, model_name)
acc_to_beat = float(model.properties["accuracy"])
except:
acc_to_beat = 0