Output

Description

Parameter Name: If you use argparse in the algorithm code to parse train_url into the data output, set the data output parameter to train_url when creating the algorithm. Set the name based on the data output parameter in your algorithm code.
The code path parameter must be the same as the data output parameter parsed in your algorithm code. Otherwise, the algorithm code cannot obtain the output path.

For example, add the train_url parameter to the code in the following way:
```
parser.add_argument('--train_url', type=str, default=None, help='test')
```
Description: customizable description of the output parameter
Obtain from: source of the output parameter. You can select Hyperparameters (default) or Environment variables.

FAQs

How Do I Store the Training Result of a ModelArts Training Job to an OBS Bucket?

Parse the output path parameter in the training code as follows. The train_url parameter can be used as the local training output path.

For more about code adaptation, see Developing a Custom Script. For more MoXing operations, see MoXing Developer Guide.

import argparse import moxing as mox # Create a parsing task.
parser = argparse.ArgumentParser(description="train mnist",                                   
                                                         formatter_class=argparse.ArgumentDefaultsHelpFormatter)  # Add the parameter.
parser.add_argument('--train_url', type=str, default='obs://obs-test/ckpt/mnist',                      
                                    help='the path model saved')  # Parse the parameter.
args, unknown = parser.parse_known_args()

When creating a custom algorithm, set the output data parameter. For details, see Creating an Algorithm.
When using an algorithm to create a training job, set the training output path to the data storage path on the cloud. For details, see Creating a Training Job.