Hay más ejemplos de AWS SDK disponibles en el GitHub repositorio de ejemplos de AWS Doc SDK.
Las traducciones son generadas a través de traducción automática. En caso de conflicto entre la traducción y la version original de inglés, prevalecerá la version en inglés.
Úselo DescribeModel con un SDK AWS
En el siguiente ejemplo de código, se muestra cómo utilizar DescribeModel.
Para obtener información, consulte Visualizar sus modelos.
- Python
-
- SDK para Python (Boto3)
-
class Models:
@staticmethod
def describe_model(lookoutvision_client, project_name, model_version):
"""
Shows the performance metrics for a trained model.
:param lookoutvision_client: A Boto3 Amazon Lookout for Vision client.
:param project_name: The name of the project that contains the desired model.
:param model_version: The version of the model.
"""
response = lookoutvision_client.describe_model(
ProjectName=project_name, ModelVersion=model_version
)
model_description = response["ModelDescription"]
print(f"\tModel version: {model_description['ModelVersion']}")
print(f"\tARN: {model_description['ModelArn']}")
if "Description" in model_description:
print(f"\tDescription: {model_description['Description']}")
print(f"\tStatus: {model_description['Status']}")
print(f"\tMessage: {model_description['StatusMessage']}")
print(f"\tCreated: {str(model_description['CreationTimestamp'])}")
if model_description["Status"] in ("TRAINED", "HOSTED"):
training_start = model_description["CreationTimestamp"]
training_end = model_description["EvaluationEndTimestamp"]
duration = training_end - training_start
print(f"\tTraining duration: {duration}")
print("\n\tPerformance metrics\n\t-------------------")
print(f"\tRecall: {model_description['Performance']['Recall']}")
print(f"\tPrecision: {model_description['Performance']['Precision']}")
print(f"\tF1: {model_description['Performance']['F1Score']}")
training_output_bucket = model_description["OutputConfig"]["S3Location"][
"Bucket"
]
prefix = model_description["OutputConfig"]["S3Location"]["Prefix"]
print(f"\tTraining output: s3://{training_output_bucket}/{prefix}")
