Model downloaded cannot be used

[cent0134]

Issue Report

Please describe the issue:

For a specific reason I can't connect to the internet on the device I'm using, so before I run dorado I need to have the mod downloaded in advance, but the mod I have downloaded never matches.After several attempts to download the mod directly, I continue to fail.Then I try to pre-run dorado on other poorly configured but network-connected devices, and interrupt the process when the moddel is matched and the download is complete. Then copy the model to the device I really want to run dorado on but can't network. But this still doesn't work.

Steps to reproduce the issue:

Run the dorado commands provided below with the files attached

Run environment:

• Dorado version:0.8.2
• Dorado command:dorado-0.8.2-linux-x64/bin/dorado basecaller hac pod5s > test1.bam --estimate-poly-a --modified-bases m6A_DRACH
• /dorado-0.8.2-linux-x64/bin/dorado basecaller /media/dell/EXTERNAL_USB/MT2/'2024-8-24 MT2'/.temp_dorado_model-8aabf2a02eeccece/rna004_130bps_hac@v5.1.0_m6A_DRACH@v1 pod5s > test1.bam --estimate-poly-a --modified-bases m6A_DRACH
• Operating system:Ubuntu
• Hardware (CPUs, Memory, GPUs):NVIDIA V100 32GB
• Source data type :POD5
• Source data location (on device or networked drive - NFS, etc.): device
• Details about data (flow cell, kit, read lengths, number of reads, total dataset size in MB/GB/TB): RNA004,2500000reads. 120GB

Logs

First, I run the following command to get a matching mod：

dorado-0.8.2-linux-x64/bin/dorado basecaller hac pod5s > test1.bam --estimate-poly-a  --modified-bases m6A_DRACH
[2024-11-19 21:34:43.982] [info] Running: "basecaller" "hac" "pod5s" "--estimate-poly-a" "--modified-bases" "m6A_DRACH"
[2024-11-19 21:34:48.516] [info]  - downloading rna004_130bps_hac@v5.1.0 with httplib
[2024-11-19 21:35:51.586] [info]  - downloading rna004_130bps_hac@v5.1.0_m6A_DRACH@v1 with httplib
[2024-11-19 21:36:04.763] [info] > Creating basecall pipeline
[2024-11-19 21:36:06.865] [info]  - BAM format does not support `U`, so RNA output files will include `T` instead of `U` for all file types.
[2024-11-19 21:36:49.992] [info] Calculating optimized batch size for GPU "Quadro P620" and model /DELL/MORE/.temp_dorado_model-b8701dd6e627943c/rna004_130bps_hac@v5.1.0. Full benchmarking will run for this device, which may take some time.

After that, I copied the mod to the device where I actually run dorado.

/media/dell/dorado-0.8.2-linux-x64/bin/dorado basecaller /media/dell/MT2/'2024-8-24 MT2'/.temp_dorado_model-8aabf2a02eeccece/rna004_130bps_hac@v5.1.0_m6A_DRACH@v1 pod5s > test1.bam --estimate-poly-a  --modified-bases m6A_DRACH
[2024-11-19 21:44:26.084] [info] Running: "basecaller" "/media/dell/MT2/2024-8-24 MT2/.temp_dorado_model-8aabf2a02eeccece/rna004_130bps_hac@v5.1.0_m6A_DRACH@v1" "pod5s" "--estimate-poly-a" "--modified-bases" "m6A_DRACH"
terminate called after throwing an instance of 'std::runtime_error'
  what():  Cannot find modification model for 'm6A_DRACH' matching simplex model: 'rna004_130bps_hac@v5.1.0_m6A_DRACH@v1'`

[HalfPhoton]

Hi @cent0134,
Please check out the documentation on how to use the model downloader and reusing models.

Let us know if that solves your issue.

Kind regards,
Rich

[malton-ont]

@cent0134,

Your command is incorrect. It should be:

dorado basecaller <path to canonical model> --modified-bases-models <path to modification model> pod5/