Deploying a trained model

Deploy a trained model to get it ready to use within IBM Maximo Visual Inspection or a different program, such as IBM Watson® Machine Learning Community Edition. Deploying a model creates a unique API endpoint based on that model for inference operations.

About this task


Models trained in IBM Maximo Visual Inspection can also be exported and deployed using the IBM Maximo Visual Inspection Edge.

To deploy the trained model, follow these steps:


  1. Click Models from the menu.
  2. Select the model you want to deploy and click Deploy.
  3. Specify a name for the model, and for models that were trained with the Optimized for speed (Tiny YOLO v2) model, choose the accelerator to deploy to. You can choose GPU or CPU.
    If custom inference scripts have been uploaded, you can optionally select Advanced deployment. These options allow you to choose a custom inference script and to specify whether inference results are saved.
    • You can save inference results to a data set without choosing a custom inference script for this deployment. You can select Advanced deployment without having custom inference scripts.
    • For image classification models trained prior to version, categories that are added as a result of an inference are listed by category UUID, rather than category name. To see category names instead, retrain the model, then run the inference again.
    For information about uploading custom inference scripts, see Preprocessing and post-processing.

    If the option to save inference results to a data set is chosen, then an existing data set must be selected for the results to be saved to. Labels generated in the inference results will be marked as "inferred" when viewing the images in the Label objects view. This can be used to validate the performance of the model.

    GPUs are used as follows:
    • Multiple Faster R-CNN and GoogLeNet models are deployed to a single GPU. IBM Maximo Visual Inspection uses packing to deploy the models. That is, the model is deployed to the GPU that has the most models deployed on it, if there is sufficient memory available on the GPU. The GPU group can be used to determine which deployed models share a GPU resource. To free up a GPU, all deployed models in a GPU group must be deleted (undeployed).
      Note: IBM Maximo Visual Inspection leaves a 500MB buffer on the GPU.
  4. Click Deploy. The Deployed Models page is displayed. When the model has been deployed, the status column displays Ready.
  5. Click the deployed model to get the API endpoint, to view details about the model, such as the owner and the accuracy, and to test other videos or images against the model.
    For information about using the API see IBM Maximo Visual Inspection API documentation.
    Note: When using the API, the smaller confidence threshold you specify, the more results are returned. If you specify 0, many, many results will be returned because there is no filter based on the confidence level of the model.
  6. If necessary, you can delete a deployed model.