Documentation for Axxon One 1.0.

Previous page Configuring detection tools  Acceleration of the neural network analytics launch on GPU Next page

It may take several minutes to launch neuroanalytics on an NVIDIA GPU after Server reboot. During this time, neuromodels are optimized for the current GPU type.

You can use caching to do this only once. Cashed optimization models are saved on the hard disk and used for next runs of analytics.

To activate caching, create the GPU_CACHE_DIR system variable and specify in its value the existing folder where the optimization models will be stored (see Appendix 10. Creating system variable).

The cache size depends on the number of neural networks used and their type, the minimum size is 70 MB.

Attention!

This function works in beta mode.

  • No labels