Files
eDNA_Stream_E01/workflow/config/main.yaml
T

23 lines
731 B
YAML

tmp_dir: "/tmp"
seed : 0
# dorado
dorado_model_dir : "../dorado_models"
hac_model_name: "dna_r10.4.1_e8.2_400bps_hac@v6.0.0"
hac_benchmarks_file: "../reference/dorado_batchsize_files/rtx4060ti_16gb_hac6.benchmarks" # add path if available
hac_min_q: 9 # empty string or 0 to disable
fast_model_name: "dna_r10.4.1_e8.2_400bps_fast@v5.2.0"
fast_benchmarks_file: "" #add path if available
fast_min_q: 8 # empty string or 0 to disable
# squiqqle dataset
pod5_dataset: "nomiss_96BC_P2I_SUP_2026"
pod5_dataset_size : 293
pod5_stride : 600 # Use 1 in x pod5 files from the ONT NO-MISS dataset to reduce dataset size / compute requirements
# custom ML
torch_env : '../envs/torch_cpu.yaml'
budget_tolerance : 0.1
reads_per_genus : 5000