Skip to content

Latest commit

 

History

History
16 lines (13 loc) · 842 Bytes

README.md

File metadata and controls

16 lines (13 loc) · 842 Bytes

Raspberry Pi Torchscript Power/Runtime Evaluation Toolkit

image

Requirements

  • Windows PC
  • Power-Z KM001 USB Power Consumption Monitor device
  • Raspberry Pi

Usage

  1. (PC) Clone jungin500/rpi-torchscript-eval-frontend
  2. (PC) Change variables on index.js and run on native or build docker image with Dockerfile
  3. (Raspberry Pi) Install this repository in $RPI_EVAL_TOOLKIT_PATH (Same path specified in index.js as RPI_EVAL_TOOLKIT_PATH)
  4. (Raspberry Pi) Prepare conda environment which can import torch and load torchscript module
  5. (PC) Open web browser and head to http://$PC_IP_ADDRESS:8080/. It will load frontend started while on Step 1~2.