😃aloha！ Thanks for your nice work.
I am wonderring how the training loop is set so that the neural codec mentioned in DCVC-HEM could realize rate adjustment in single model. I am kind of a newbie in this area, after reading this paper DCVC-HEM and the test code provided by this repo, I have no idea about this. The paper says the training strategy is set as same as TCM, so I read the referenced paper TCM to figure out how to realize a multi-stage training. But after reading the paper I still don't know about this.
So could you please provide a little help or share the training code?
😃aloha！ Thanks for your nice work. I am wonderring how the training loop is set so that the neural codec mentioned in
DCVC-HEMcould realize rate adjustment in single model. I am kind of a newbie in this area, after reading this paperDCVC-HEMand the test code provided by this repo, I have no idea about this. The paper says the training strategy is set as same asTCM, so I read the referenced paperTCMto figure out how to realize a multi-stage training. But after reading the paper I still don't know about this. So could you please provide a little help or share the training code?