Debugging a Model

After the training is complete, create a local model, debug the model locally, and deploy the model on ModelArts.

Sample Code

In ModelArts notebook, you do not need to enter authentication parameters for session authentication. For details about session authentication of other development environments, see Session Authentication.

Save the custom inference file and model configuration file to the directory storing the model file generated during training. If the model generated during training is stored in /home/ma-user/work/tensorflow_mlp_mnist_local_mode/train/model/, the inference file customize_service.py and model configuration file config.json are also stored in this directory.

Create a Conda virtual environment for running models.

      
           from modelarts.environment import Environment
from modelarts.environment.conda_env import CondaDependencies

env = Environment("tensorflow_mlp_mnist")
cd = CondaDependencies.create(pip_packages=["tensorflow==1.13.1", "Pillow>=8.0.1"],
                              conda_packages=["python=3.6.2"])
env.conda = cd

Create a local model.

      
           from modelarts.session import Session
from modelarts.model import Model

session = Session()
src_local_path = "/home/ma-user/work/tensorflow_mlp_mnist_local_mode/train/"
model = Model(session,
              publish=False,
              source_location_type="LOCAL_SOURCE",         # Type of the model file location
              source_location=src_local_path + 'model',    # Location of the model file
              environment=env,
              model_version='1.0.1',
              model_type='TensorFlow',                     # AI framework used by the model
              model_algorithm="image_classification",
              model_name="tensorflow_mlp_mnist_local_infer")

After a local model is created, you can deploy it as a local service.

Call the API to publish the model.

      
           model.publish_model(obs_location=obs_location)

After the obs_location parameter is specified, the local model file is uploaded to this directory. This parameter can be omitted. See the following example:

      
           model.publish_model()

The model file is uploaded to the directory whose name ends with the current timestamp in the default OBS bucket. The directory is displayed after the command is executed:

      
           Successfully upload file /home/ma-user/work/tensorflow_mlp_mnist_local_mode/train/model to OBS modelarts-cn-north-4-08aae033/model-0107-224502

Parameters

**Table 1** Parameters for creating a model
Parameter	Mandatory	Type	Description
session	Yes	Object	Session object. For details about the initialization method, see Session Authentication.
model_name	No	String	Name of a model that consists of 1 to 64 characters and must start with a letter. Only letters, digits, underscores (_), and hyphens (-) are allowed. If this parameter is not specified, the system automatically generates a model name.
model_version	Yes	String	Model version in the format of Digit.Digit.Digit. The value range of the digits is [1, 99]. The version number cannot start with 0, for example, 01.01.01.
publish	No	Bool	Whether to publish a model. The options are as follows: True: Publish the model. (Default value) False: Do not publish the model. Create a local model, which can be used to debug related code.
source_location_type	No	String	Model location type. The options are as follows: OBS_SOURCE: OBS path. (Default value) LOCAL_SOURCE: local path.
source_location	Yes	String	Path (parent directory) of the model file If source_location_type is set to OBS_SOURCE, the model file path is an OBS path in the format of /obs_bucketname/.../model_file_parent_dir/. If source_location_type is set to LOCAL_SOURCE, the model file path is a local path in the format of /local_path/.../model_file_parent_dir/.
environment	No	Environment instance	Environment required for normal model running, such as the Python or TensorFlow version For details, see Table 2.
source_job_id	No	String	ID of the source training job. If the model is generated from a training job, specify this parameter for source tracing. If the model is imported from a third-party meta model, leave this parameter blank. By default, this parameter is left blank.
source_job_version	No	String	Version of the source training job. If the model is generated from a training job, specify this parameter for source tracing. If the model is imported from a third-party meta model, leave this parameter blank. By default, this parameter is left blank.
source_type	No	String	Model source type. The value can only be auto, which indicates an ExeML model (model download is not allowed). If the model is deployed via a training job, leave this parameter blank. By default, this parameter is left blank.
model_type	Yes	String	Model type. The value can be TensorFlow, MXNet, Spark_MLlib, Scikit_Learn, XGBoost, MindSpore, Image, or PyTorch.
model_algorithm	No	String	Model algorithm. If the algorithm has been configured in the model configuration file, this parameter can be left blank. Possible options are predict_analysis, object_detection, and image_classification.
description	No	String	Model description, which contains a maximum of 100 characters and cannot contain the following special characters: !<>=&'"
execution_code	No	String	OBS path to the execution script. The inference script must be stored in the model directory in the path where the model is located. For details, see the source_location parameter. The script name is fixed to customize_service.py.
input_params	No	params array	List of input parameters for model inference. By default, this parameter is left blank. If the apis information has been configured in the model configuration file, you do not need to set this parameter. The backend automatically reads the input parameters from the apis field in the configuration file.
output_params	No	params array	List of output parameters for model inference. By default, this parameter is left blank. If the apis information has been configured in the model configuration file, you do not need to set this parameter. The backend automatically reads the output parameters from the apis field in the configuration file.
dependencies	No	dependency array	Dependency package required for running the code and model. By default, this parameter is left blank. If the dependencies information has been configured in the model configuration file, you do not need to set this parameter. The backend automatically reads the dependencies to be installed from the dependencies field in the configuration file.
apis	No	String	List of inference APIs provided by a model. By default, this parameter is left blank. If the apis information has been configured in the model configuration file, you do not need to set this parameter. The backend automatically reads the configured inference API information from the apis field in the configuration file.

**Table 2** Environment parameters
Parameter	Mandatory	Type	Description
name	Yes	String	Environment name
conda	No	CondaDependencies	Conda environment. For details, see Table 3.

**Table 3** CondaDependencies parameters
Parameter	Mandatory	Type	Description
channels	No	List	Source for downloading the Python package
pip_packages	No	List	Python package required by the Conda virtual environment, such as TensorFlow and Pillow
conda_packages	No	List	Conda package required by the Conda virtual environment, for example, a specified Python version

**Table 4** **params** parameters
Parameter	Mandatory	Type	Description
url	Yes	String	Request path of a model inference API
param_name	Yes	String	Parameter name, which contains a maximum of 64 characters
param_type	Yes	String	Basic parameter types of JSON schema, including string, object, array, boolean, number, and integer
min	No	Double	This parameter is optional when param_type is set to int or float. By default, this parameter is left blank.
max	No	Double	This parameter is optional when param_type is set to int or float. By default, this parameter is left blank.
param_desc	No	String	Parameter description, which contains a maximum of 100 characters. By default, this parameter is left blank.

**Table 5** **dependency** parameters
Parameter	Mandatory	Type	Description
installer	Yes	String	Installation mode, which can only be pip
packages	Yes	package array	Collection of dependency packages

**Table 6** **package** parameters
Parameter	Mandatory	Type	Description
package_name	Yes	String	Name of a dependency package
package_version	No	String	Version of a dependency package
restraint	No	String	Version filtering condition. This parameter is mandatory only when package_version is available. Options: EXACT: a specified version ATLEAST: not earlier than the specified version ATMOST: not later than the specified version