mirror of
https://jihulab.com/mr-one/openpilot.git
synced 2026-10-27 21:13:42 +08:00
Comma Device
f1db2eb5aa
Downloaded/tinygrad-runner models are compiled with tinygrad e837e367a (the pinned tinygrad_ref in the model catalogs). The old 66ee3cfb4 runtime crashes rendering their uops (uop/render.py TypeError), so the device must ship this tinygrad for the new modeld_v2 pipeline.
Each model should be a clean single file. They are imported from the top level `models` directory It should be capable of loading weights from the reference imp. We will focus on these 5 models: # Resnet50-v1.5 (classic) -- 8.2 GOPS/input # Retinanet # 3D UNET (upconvs) # RNNT # BERT-large (transformer) They are used in both the training and inference benchmark: https://mlcommons.org/en/training-normal-21/ https://mlcommons.org/en/inference-edge-30/ And we will submit to both. NOTE: we are Edge since we don't have ECC RAM