Frequently Asked Questions

I get an error using the AirPack Docker container on the AIR-T

A common error when trying to do this is exec user process caused exec format error.

This happens because AirPack was not designed run on the AIR-T because the AIR-T has limited GPU compute capability for training. If you look at the requirements in the file, it will let you know what is required to run the software and there is also additional information on our website:

The training of the neural network will need to be performed on a computer that supports TensorFlow. While the AIR-T does contain a GPU, training on it will take significantly more time than on a larger PCIe-based GPU. As a result, it is recommended to train on a desktop computer or server with such a card installed. Feel free to contact us for recommendations on suitable cards for training.

AirPack runs on a training computer and produces a trained model as the output. That trained model may be executed on the AIR-T, but the AIR-T is not meant to execute the training step.

Last update: February 24, 2020