search

hongzimao / pensieve

Neural Adaptive Video Streaming with Pensieve (SIGCOMM '17)
http://web.mit.edu/pensieve/
MIT License
516 stars 279 forks source link

cross-validation curve #144

Open ahmad-hl opened 2 years ago

ahmad-hl commented 2 years ago

I was trying to monitor the TD_loss and rewards using tensorboard by launching it using this command: sim> tensorboard logdir=./results However, I couldn't see the validation curve with the testing. In other words, there is no summary writer in rl-test.py that write the summary [TD_loss, entropy, and reward] of testing! I can only see the summary [TD_loss, entropy, and reward] of learning from sim/multi-agent.py. How can I enable and show the cross-validation curve to monitor the model convergence?

ahmad-hl commented 2 years ago

Hi Hongzi,

I have updated the repo to support both the training & testing curves on tensorboard. It is now available at this repo: https://github.com/ahmad-hl/pensieve-py38

Should I fork your repo then provide the update in the forked one, or it is okay to keep it like this?