-
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…
-
Torchtune could provide a recipe to do HPO, where the user provides a config, the recipe, eval dataset, params to sweep and budget.
I just played with optimizer. Our default in lr 3e-4. I tried 3e-…
-
Thank you very much for sharing, but when I extracted the data from the demo files, I found that the `res18_x128_vert_xgz.pth` required to run the demo was not present in the data.
-
Environment: DEV and TEST with UDS v8.0.0
For collections not configured for DAAC archival (which is most), granules_to_es lambda fails which results in messages not clearing from granules_to_es_qu…
-
GitLab support "Secret token" when sending webhooks:
Used to validate received payloads. Sent with the request in the X-Gitlab-Token HTTP header.
Adding this support into ucascade would make it a lo…
-
Could u please add a tableofcontent in github?
Thanks a lot!
-
We sort the array using multiple passes. After the first pass, the maximum element goes to end (its correct position). Same way, after second pass, the second largest element goes to second last posit…
-
-
Make it possible to start certain or all weapons with a specified amount of ammo when you pick them up.
Like for example, laptop gun = 400 ammo
If you make it so AI can use laptop gun's seconda…
-
### Describe the improvement you'd like to request
Add tests for the use of keywords in Cedar and JSON formats to make sure that 1) the results are what we expect and 2) the parser error messages mak…