rnaglib.tasks.ResidueClassificationTask

class rnaglib.tasks.ResidueClassificationTask(additional_metadata=None, **kwargs)

Classification task at the residue level.

Each residue (nucleotide) in the RNA is classified independently.

__init__(additional_metadata=None, **kwargs)

Methods

__init__([additional_metadata])
add_feature(feature[, feature_level, is_input])	Add a feature to the dataset.
add_representation(representation)	Add a representation transform to the dataset.
add_rna_to_building_list(all_rnas, rna)	Add an RNA to the building list.
compute_distances()	Compute similarity distances between RNAs in the dataset.
compute_metrics(all_preds, all_probs, all_labels)	Compute classification metrics aggregated across all predictions.
compute_one_metric(preds, probs, labels)	Compute classification metrics for a single set of predictions.
create_dataset_from_list(rnas)	Compute an RNADataset object from the lists touched in add_rna_to_building_list.
describe()	Get description of task dataset.
dummy_inference()	Run dummy inference on the test dataset.
evaluate(model, loader)	Evaluate model performance on a dataset.
from_scratch(size_thresholds)	Create task dataset from scratch.
from_zenodo()	Download the task dataset from Zenodo and load it.
get_split_datasets([recompute])	Get train, validation, and test datasets.
get_split_loaders([recompute])	Get train, validation, and test dataloaders.
get_task_vars()	Define a FeaturesComputer object to set which input and output variables will be used in the task.
init_metadata([additional_metadata])	Initialize dictionary to hold key/value pairs to self.metadata.
load()	Load dataset and splits from disk.
post_process()	Apply post-processing steps to remove redundancy.
process()	Tasks must implement this method.
remove_redundancy()	Remove redundant RNAs from the dataset based on similarity.
remove_representation(representation_name)	Remove a representation transform from the dataset.
set_datasets([recompute])	Set the train, val and test datasets.
set_loaders([recompute])	Set the dataloader properties.
split(dataset)	Calls the splitter and returns train, val, test splits.
to_csv(path)	Write a single CSV with all task data.
write()	Save task data and splits to root.

Attributes

default_splitter	The splitter used if no other splitter is specified.
dummy_model	Get a dummy model for testing purposes.
task_id	Task hash is a hash of all RNA ids and node IDs in the dataset.