Yingru Li
Yingru Li
> Hi @Hsintao, data for VOC needs to be in COCO's JSON format. Read more about preparing data [here](https://github.com/adityaarun1/Detectron.pytorch#data-preparation). > > P.S.: To answer the previous question, the setting for...
> I suggest that you refer to this paper: IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures. > I do have a plan add the algorithm to this platform....
> Can running and training be separated? For example, we deploy on the cloud, send data to the cloud for training, and issue policies to local hosts intermittently or in...
> Unfortunately I have not seen this error before. How did you install the environment? Are you on the latest master? yes, I cloned the latest master. and I just...
> Remark: In our current implementations of theoretically grounded algorithms (`UCBVI`, `OptQL`, `AdaptiveQL`, `RSKernelUCBVI`, `LSVI-UCB`), we often need some tricks to make them more "practical" (i.e. not have a linear...
> @yix081 Could you post your modifications here? Thank you!
@rPortelas @yix081 Thank you! Very interested in the reproducibility issue. Is there anything to share about the debugging in the current reanalyze.py?
Any update on this issue? Thanks
Thanks for the discussion! Any follow-up message so far?
this code might be based on pytorch 0.3.0