mirror of
https://github.com/RosettaCommons/foundry.git
synced 2026-06-04 13:24:22 +08:00
5e7b739ed3
* Add initial files * add files * Move projects.aa_design -> rfd3 * Make format * Delete test files * Add configs * Mc * Fixed tests * remove test files
39 lines
1.2 KiB
YAML
39 lines
1.2 KiB
YAML
defaults:
|
|
- pdb/base_transform_args@monomer_distillation
|
|
- _self_
|
|
|
|
monomer_distillation:
|
|
dataset:
|
|
_target_: atomworks.ml.datasets.datasets.StructuralDatasetWrapper
|
|
save_failed_examples_to_dir: ${paths.data.failed_examples_dir}
|
|
# Explicitly do not load from cache.
|
|
# Dataset too big, and structures are small
|
|
cif_parser_args:
|
|
cache_dir: null
|
|
load_from_cache: False
|
|
save_to_cache: False
|
|
|
|
# metadata dataset
|
|
dataset:
|
|
_target_: atomworks.ml.datasets.datasets.PandasDataset
|
|
name: af2fb_distillation
|
|
id_column: example_id
|
|
data: ${paths.data.monomer_distillation_parquet_dir}/af2_distillation_facebook.parquet
|
|
columns_to_load:
|
|
- example_id
|
|
- path
|
|
|
|
# metadata parser
|
|
dataset_parser:
|
|
_target_: atomworks.ml.datasets.parsers.GenericDFParser
|
|
pn_unit_iid_colnames: null
|
|
|
|
transform:
|
|
_target_: ${datasets.pipeline_target}
|
|
is_inference: False
|
|
# protein_msa_dirs: [{"dir": "${paths.data.monomer_distillation_data_dir}/msa", "extension": ".a3m", "directory_depth": 2}]
|
|
# rna_msa_dirs: []
|
|
crop_contiguous_probability: 0.25
|
|
crop_spatial_probability: 0.75
|
|
b_factor_min: 70
|