How to train in lower bitrates ( < 0.14)?

Justin-Tan / high-fidelity-generative-compression

Pytorch implementation of High-Fidelity Generative Image Compression + Routines for neural image compression

Apache License 2.0

411 stars 77 forks source link

How to train in lower bitrates ( < 0.14)? #12

Closed QLaHPD closed 3 years ago

QLaHPD commented 3 years ago

What is needed to change in the code to train with a different bitrate?

Justin-Tan commented 3 years ago

Assuming you are working in the low regime, which is the default, change the value for low in the target_rate_map dict in default_config.py to whatever bitrate you want.