Kevin Pedro

I have a question about converting from `hep_ml` BDTs to `sklearn` BDTs. I am trying to use the "exporting to TMVA" method via [koza4ok](https://github.com/yuraic/koza4ok), and it works with a few...

Thanks @guitargeek, I'm able to run all the benchmarks now and I get similar, though not exactly the same, behavior on my Intel CPU. I also observe that the results...

I'm just going to summarize the details of my custom reader here, since it turns out not to be faster. (At one time, I had observed it to be faster...

If I may make a suggestion: it would be especially useful to have the download path as a separate box from the mask (a la the old DTA). This would,...

Fair enough. As a lower-priority item, maybe an option for a separate box (disabled by default) could eventually be added.

Unfortunately, this is a change in Firefox; they have limited the APIs available for WebExtensions. See my comment at https://github.com/downthemall/downthemall/issues/2#issuecomment-526801494 for a user-side workaround (and note that you can put...

> You potentially added 96908 lines to the logs This should be fixed...

Appears to be the result of an actual bug; this message is repeated many times in Phase 2 step 3 workflow logs: ``` %MSG-w PFClusterAlgo: PFClusterProducer:particleFlowClusterHGCal 08-Aug-2022 16:07:53 CEST Run:...

Just wanted to comment that support for deadlines (from the client side) is very important for the async callback mode (which seems to be an active area of development). Even...

@lmoneta in CMS we are using THnSparse for a fairly large-scale analysis, so any suggestions of viable fixes or workarounds that we can easily implement would be very appreciated!