Skip to content

Latest commit

 

History

History
15 lines (10 loc) · 737 Bytes

README.rst

File metadata and controls

15 lines (10 loc) · 737 Bytes

Example applications

These are 5 example models; in order of complexity

  • app_no_flash - a single model, no flash memory used. This is the fastest but most pressure on internal memory.
  • app_flash_single_model - a single model, with learned parameters in flash memory. This removes a lot of pressure on internal memory.
  • app_flash_two_models - two models, with learned parameters in flash memory.
  • app_flash_two_models_one_arena - two models, with learned parameters in flash memory. The models share a single tensor arena (scratch memory).