Troubleshooting

Download Server Logs

To know more about an error, download the server logs by clicking the box with three dots next to the Version field and pressing Download Log:

download_log-b.png

A .txt file with the server logs is downloaded. Use the logs to investigate the problems and manually run tools to debug the problem by entering the Docker* container. For more information, go to Enter Docker Container with DL Workbench.

Most Frequent Mistakes

Incompatible Model and Dataset

troubleshooting_model_dataset_01-b.png

This error appears due to model and dataset type incompatibility.

calibrator_usage-01-b.png

Also, check that you do not select a VOC Object-Detection dataset for a Classification model, or an ImageNet Classification dataset for an Object-Detection model.

Range of Inferences Configuration

troubleshooting_min_max-b.png

When configuring numbers of streams and batches to run a range of inferences, make sure minimum values are not greater than maximum ones, as well as the number of steps is not greater than the difference between maximum and minimum values.

Wrong Archive

An error shown below may appear due to loading a wrong archive:

troubleshooting_wrong_archive-b.png

Check the archive with a model or a dataset. This model must contain two files: .xml and .bin.

Open Model Zoo Models Do Not Get Imported

troubleshooting_omz-b.png

If you cannot import models from the Open Model Zoo, you may need to specify your proxy settings when running a Docker container. For details, refer to Install from Docker Hub*.