search

NVIDIA-AI-IOT / jetson_benchmarks

Jetson Benchmark
MIT License
372 stars 71 forks source link

fix the bug : not work on jp>4.6 #29

Closed wanghr323 closed 2 years ago

wanghr323 commented 2 years ago
  1. tensorrt8 instead "end-to-end" metric with "Host lantency", use py-tensorrt to get the version of tensorrt
  2. for jp>5.0(rel-34), the pwn_fan changed
  3. NX add a new power-mode (nvpmodel 8) as the MAXN power mode, update this in README.md
  4. write the convert-log to file as well which can help to debug
  5. fix warning: jetson_benchmarks/utils/read_write_data.py:126: SyntaxWarning: "is" with a literal. Did you mean "=="?