malton-ont

Hi @fayora. Can you try the workaround from [this comment](https://github.com/nanoporetech/dorado/issues/579#issuecomment-1904282293)?

Hi @fayora, that's great to hear. No, this should have no effect on the results.

`DeviceStats::requested_bytes` was introduced in libtorch 2.0. Are you linking against your own version of libtorch rather than the version that the dorado configuration process downloads?

@avilella - the modification model you are running (`5mCG_5hmCG`) is selecting only `CG` contexts. This means dorado doesn't attempt to analyse any `C` bases that aren't immediately followed by a...

> Modbase calling in Dorado runs concurrently with basecalling, please refer to the [Dorado documentation on this](https://github.com/nanoporetech/dorado?tab=readme-ov-file#modified-basecalling) for information on how to run it. Just to clarify this - modbase...

Hi @Afollet, The trace showing scores only for the front barcode is slightly misleading here - for double ended barcodes we do check against both, but the trace output only...

Hi @Afollet. Yes, we can improve that logging in a future release. If you don't want to wait you can compile dorado for yourself and add it in around [here](https://github.com/nanoporetech/dorado/blob/release-v0.5.3/dorado/demux/BarcodeClassifier.cpp#L629)...

Hi @sjaenick, Thanks for your report. The relevant lines here are: ``` [2024-02-27 12:39:17.352] [warning] Auto batchsize detection failed. Less than 1GB GPU memory available. [2024-02-27 12:39:17.355] [info] - set...

@DCossey, for out of memory issues can you try [this](https://github.com/nanoporetech/dorado/issues/579#issuecomment-1904282293) solution?

Hi @JWDebler, This is on hold at the moment while we work on other priorities.