search

DerrickXuNu / OpenCOOD

[ICRA 2022] An opensource framework for cooperative detection. Official implementation for OPV2V.
https://mobility-lab.seas.ucla.edu/opv2v/
Other
644 stars 99 forks source link

The log file has generated garbled characters #116

Closed githigher closed 9 months ago

githigher commented 9 months ago

Why did my training log file 'events. out. tfevents. *' generate garbled characters? In your code, there are several sentences that write the loss value to the log file:

  1. writer.add_scalar('Validate_Loss', valid_ave_loss, epoch)
  2. writer.add_scalar('Regression_loss', reg_loss.item(), epochbatch_len + batch_id) writer.add_scalar('Confidence_loss', conf_loss.item(), epochbatch_len + batch_id)

But when I opened the log file, except for "Validate-Loss", "Registration_loss", and "Confidence_loss", all other values were garbled. I have tried many encoding methods for reading, but the result still doesn't work.