Deploy Models

You are currently viewing the documentation for the latest version (3.0.0). To access a different version, click the "Switch version" button located in the upper-right corner of the page.

■ If you are not sure which version of the product you are currently using, please feel free to contact Mech-Mind Technical Support.

Models trained with Mech-DLK can be used in applications like 3D robot guidance, 3D measurement and inspection. This document outlines the usage of the models in Mech-Vision, Mech-MSR, and Mech-DLK SDK.

Deploy Models in Mech-Vision

Usage Instructions

The exported model can be used in Mech-Vision. For more information, see Deploy Models – Instance Segmentation for 3D Robot Guidance.

Compatibility

Version

It is recommended to use models exported from Mech-DLK 3.0.0 in conjunction with Mech-Vision 2.2.0.

Model Package

Model packages (including single model packages and multi-model packages) exported from Mech-DLK 3.0.0 can only be used in Mech-Vision 2.2.0, not versions earlier than Mech-Vision 2.2.0.

Single model packages and multi-model packages are defined as follows:

  • Single-model package: A model package that contains one and only one algorithm module.

  • Multi-model package: A model package that contains multiple algorithm modules (i.e., tree-structured algorithm modules), where the modules are combined in serial, parallel, or hybrid serial-parallel modes.

For detailed information about the compatibility of each module, please refer to Deep Learning Compatibility.

Use Models in Mech-MSR

Model packages (including single model packages and multi-model packages) exported from Mech-DLK 3.0.0 can only be used in Mech-MSR 2.2.1, not versions earlier than Mech-MSR 2.2.1.

  • The Picking Anything V2 and Object-Bin Segmentation model packages cannot be used in Mech-MSR.

  • Model packages of other modules support inference through the Deep Learning Model Package Inference Step in Mech-MSR.

For detailed information about the compatibility of each module, please refer to Deep Learning Compatibility.

Use Models in Mech-DLK SDK

Mech-DLK SDK is a software development kit specifically designed to be used with Mech-DLK. Its main purpose is to help developers easily do deep learning inference in their software systems. With this kit, developers can rapidly deploy deep learning models trained in Mech-DLK and flexibly integrate deep learning functionality into their own applications without reliance on Mech-Vision or Mech-MSR.

Usage Instructions

If you need to use Mech-DLK SDK, see Mech-DLK SDK User Manual for details.

You can also visit Mech-Mind’s Download Center to get related resources.

Compatibility

Model packages exported from Mech-DLK 2.5.4 or above can be used in Mech-DLK SDK 3.0.0.

  • The Picking Anything V2 and Object-Bin Segmentation model packages exported from Mech-DLK 3.0.0 cannot be used in Mech-DLK SDK.

  • Other model packages exported from Mech-DLK 3.0.0 can only be used in Mech-DLK SDK 3.0.0, not on versions earlier than Mech-DLK SDK 3.0.0.

It is recommended that the latest versions of both Mech-DLK and Mech-DLK SDK be used to ensure optimal compatibility and functionality.

Is this page helpful?

You can give a feedback in any of the following ways:

We Value Your Privacy

We use cookies to provide you with the best possible experience on our website. By continuing to use the site, you acknowledge that you agree to the use of cookies. If you decline, a single cookie will be used to ensure you're not tracked or remembered when you visit this website.

  • Preference
  • I accept
  • I don't accept