Manage a model quantization job

Manage a model quantization job

Model quantization provides you with a dedicated environment to develop and experiment with your AI models. After creating a model quantization job, follow these steps to seamlessly manage your model training job:

Step 1: Accessing Model Quantization

  • Dashboard: From the Greennode AI Platform dashboard, locate the "Model quantization" section.
  • Model Quantization List: You'll see a list of your existing model quantization, including their names, creation dates, status (e.g., running, stopped), and configuration.

Step 2: Monitoring a Model Quantization

After creating your model quantization, you can access it by clicking on its name. This will take you directly to the model quantization detail interface where you can begin monitoring it.
The model quantization job details page provides comprehensive information about your quantization job's progress, resource usage, and logs. It's divided into three main sections:

General Information

Use this section to quickly assess the overall status and progress of your quantization job. The configuration details are helpful for reviewing your quantization setup and troubleshooting any issues.
  • Job Name: A unique identifier for your quantization job.
  • Status: Indicates the current state of the job (e.g., Queued, Running, Completed, Failed).
  • Duration: The total time the job ran (or is currently running).
  • Configuration: Details about the model architecture, hyperparameters, and other settings used for quantization.

Usage Information

Use this tab to verify that your datasets are correctly mounted and monitor quantization logs to track progress, identify potential issues, and debug errors. System logs can be helpful for troubleshooting infrastructure-related problems. This section has two tabs:

  1. Data Volume: Displays information about the datasets or source codes you've mounted to your training job from external sources (e.g., S3 buckets). This includes the source location, mount path, etc.
  2. Logs: A real-time stream of logs generated during the quantization process. These logs contain information about the model's quantization progress, metrics, and any errors that occur.

Step 3: Stopping Quantization Job

Stop quantization
If you need to pause your work or save on resources, you can stop your quantization job instance. Select the quantization job you wish to stop and click the "Stop" button. This will halt the instance until you restart it again.
  1. Locate the quantization job you want to stop in the instance list.
  2. Click the "Stop" button.
  3. Wait for the quantization status to change to "Stopped."
NotesImportant Considerations:
  • Billing: Training job instances incur charges while they are running. Be sure to stop instances when you are not actively using them to avoid unnecessary costs.
  • Data Persistence: Data stored on the instance's local storage will not persist when the instance is stopped. Therefore, remember to store your result as a checkpoint & output data on another space for backup solution (e.g., s3 bucket) when creating the model training job.

Step 4: Delete Quantization Job

When a quantization job is no longer needed, you can delete it to free up resources. To delete an instance, choose the quantization job and click the "Delete" button. A confirmation dialog will appear to ensure you do not accidentally delete the wrong instance. Please note that once a quantization is deleted, it cannot be recovered.

    • Related Articles

    • Manage a model tuning job

      Model tuning, also known as hyperparameter optimization, is the process of adjusting the hyperparameters of a machine learning model to improve its performance. Hyperparameters are settings that determine the learning process of a model and are not ...

    • Create a quantization job

      To create a quantization job, you'll need to provide the following information: Steps to Create a Quantization Tuning Job Access the Quantization Job Creation Interface: Use the provider's platform through the url: . Fill in the Input Parameters: ...

    • Manage a Model Registry

      The Model Registry serves as a centralized repository to track, organize, and manage your AI/ML models. It ensures reproducibility, versioning, and simplifies deployment. Follow this guide to understand how to manage your model registry and prepare ...

    • Manage a model endpoint

      This guide will walk you through the key features and steps involved in deploying your models, optimizing costs through undeployment, and removing endpoints when they are no longer needed. After creating a model endpoint, follow these steps to ...

    • Create a tuning job

      To create a supervised tuning job, you'll need to provide the following information: Steps to Create a Supervised Tuning Job Access the Tuning Job Creation Interface: Use the provider's platform through the url: . Fill in the Input Parameters: ...