-
Hi, thanks for the great work and sharing your wandb training logs! After analysing the plots, I have some questions regarding the upcycling experiment done for OLMoE and would greatly appreciate if y…
-
Do VID benchmarks with many CPU cores:
1. Try (locally) to identify a trend when measuring the duration of disperse while increasing the number of available CPU cores.
2. Spin up AWS machine with 32…
-
-
Implement gaussian+low-E side tail for MJD partitions.
Idea: add a key in the partition.json (under `fit_group`?) for specifying the peak shape to use for partitions listed in that file, eg `"signal_…
-
Hello! In your four detection experiments, I noticed a significant difference between the reported result and the test result for `mask_rcnn_swin_small`. The reported result is 44.2 mAP, but the teste…
-
Fantastic work!
Could you please tell me when the new model and checkpoints will be released? My collaborators and I are looking forward to taking advantages of this great work :)
Thank you very muc…
-
I hope this message finds you well.
I came across your impressive project on GitHub and noticed the related dataset used in your work. I am currently working on a similar research and would greatly…
-
Hi ,what's your data split for few-shot experiments? For example, in one-shot or two-shot setting, how to split training/validation set?
-
Hello, I’ve been following your work recently. Based on the configurations in your repo, it seems that the reward queries for REBEL are twice as for DDPO, since REBEL uses two sampling traces per batc…
-
I'm gearing up to do some fine-tuning experiments with Open-Sora in my free time. Could you give me a quick rundown on what makes ideal training data for a small-scale setup?
Specifically, I'm wond…