-
Notifications
You must be signed in to change notification settings - Fork 53
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Duplex basecalling is impractically slow #98
Comments
Hi, it’s possible that this is caused by a low pairing rate. Dorado v0.2.1
introduces performance improvements to Dupelx, could you try this?
…On Wed, 15 Feb 2023 at 23:36, Hasindu Gamaarachchi ***@***.***> wrote:
The command I ran is: dorado duplex ***@***.***
reads/ --pairs pairs_from_bam_hg2/pair_ids_filtered.txt
Now 19 hours, but only 120,000 reads are done, on 4xTesla V100. For 1.2M
pairs, it will be a few days. GPUs are consistently running at 80-95%
utilisation in nvidia-smi. Is it normal to be this slow?
—
Reply to this email directly, view it on GitHub
<#98>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AALYB7IYK7P57GYRCDECFDLWXVR63ANCNFSM6AAAAAAU5O6T7M>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Keeps getting this with the newest dorado
|
Hi @hasindu2008 , is the model |
Now the time for duplex is good with 0.2.1! Took only around 6 hours. |
That's great! We are working on further performance improvements so expect that 6 hours to go down further. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The command I ran is:
dorado duplex dna_r10.4.1_e8.2_400bps_sup@v4.0.0 reads/ --pairs pairs_from_bam_hg2/pair_ids_filtered.txt
Now 19 hours, but only 120,000 reads are done, on 4xTesla V100. For 1.2M pairs, it will be a few days. GPUs are consistently running at 80-95% utilisation in nvidia-smi. Is it normal to be this slow?
The text was updated successfully, but these errors were encountered: