search

nanoporetech / bonito

A PyTorch Basecaller for Oxford Nanopore Reads
https://nanoporetech.com/
Other
389 stars 120 forks source link

Help us on Bonito expected run time #296

Open Saranya-nanobiophotonics-lab opened 2 years ago

Saranya-nanobiophotonics-lab commented 2 years ago

Hi @iiSeymour We are using NVIDIA V100 GPU. We have created an virtual environment to run bonito. For test training run, we have used a fast 5 file of size 152MB (corresponding to ~2750reads). We used the following command "bonito basecaller dna_r10.4_e8.1_hac@v3.4 --save-ctc --reference ref.mmi fast5_sub > ctc-data/basecalls.sam" The code is running for over 24 hrs with 0% preprocessing of reads. PFA and please comment.

image

Saranya-nanobiophotonics-lab commented 2 years ago

Could someone please comment on this issue.

iiSeymour commented 2 years ago

Hi @Saranya-nanobiophotonics-lab this looks like it's stalled with an I/O issue and not specifically a bonito error.

Is your data on a local disk or a network drive here?

Saranya-nanobiophotonics-lab commented 1 year ago

Hi @iiSeymour The data is in local disk. I had successfully base-called and performed model training. Further, I tried base-calling using the new model. But, I get poly G as data output.