Convert indices to sequences.

Return binary predictions with a threshold. Parameters ---------- threshold : float, optional Decision threshold

Return binary predictions using the specified probability threshold.

Return thresholded predictions along with confidence scores.

Return thresholded predictions along with confidence scores.

Return thresholded predictions along with confidence scores.

Convenience inference wrapper producing final predictions. Behavior depends on ``task_name``: - ``multilabel_classification``: thresh

Perform inference for masked language modeling, decoding predictions to sequences. Args: sequence_or_inputs: Input seque

Perform inference for token-level regression, excluding special tokens. Args: sequence_or_inputs: Input sequences or pre

Perform inference for sequence-level regression. Args: sequence_or_inputs: Input sequences or pre-processed inputs

Perform inference for matrix regression. Args: sequence_or_inputs: Input sequences or pre-processed inputs *

Perform inference for matrix classification. Args: sequence_or_inputs: Input sequences or pre-processed inputs

This method provides processed, human-readable sequence-level predictions. It converts logits to class labels and provides confidence

Performs multi-label inference with human-readable output. It converts logits to binary labels and provides confidence scores.

Perform forward pass and return the last hidden state from the base model. This method provides access to intermediate representatio

This method loads a complete pipeline including the model, tokenizer, datasets, and trainer configuration. If the pipeline doesn't ex

This function automatically downloads benchmark datasets if they don't exist locally, loads their configurations, and initializes train/valid

Load a model and its tokenizer from Hugging Face Hub or local path. This method loads both the model and tokenizer, places them on t

Binary cross-entropy loss with logits already in probability space. Notes ----- This legacy baseline uses ``BCELoss`` assumin

Binary cross-entropy loss on probabilities.

Compute task-appropriate training loss.

Compute the loss for masked language modeling. Args: logits (torch.Tensor): Model predictions [batch_size, seq_len, voca

Compute the loss for token-level regression. Args: logits (torch.Tensor): Model predictions labels (torch.Te

Compute the loss for sequence-level regression. Args: logits (torch.Tensor): Model predictions labels (torch

Compute the loss for matrix regression. Args: logits (torch.Tensor): Model predictions labels (torch.Tensor)

Compute the loss for matrix classification. Args: logits (torch.Tensor): Model predictions labels (torch.Ten

Calculates the cross-entropy loss for token classification. This method computes the cross-entropy loss between the predicted

This method computes the cross-entropy loss between the predicted logits and the ground truth labels. Args: logi

Calculates the binary cross-entropy loss for multi-label classification. Args: logits (torch.Tensor): Predicted logits f

Compute Mean Column Root Mean Square Error (MCRMSE). MCRMSE is a multi-target regression metric that computes the RMSE for each target c

Compute Mean Column Root Mean Square Error (MCRMSE). MCRMSE is a multi-target regression metric that computes the RMSE for each target co

Create a minimal mock binary classification dataset. Based on translation efficiency prediction pattern. Limited to 10 samples per split

Create minimal mock dataset for CLI testing

Create a minimal mock multi-label classification dataset. Based on TFB prediction (DeepSEA) pattern with 919 labels. Limited to 10 sample

Load model for structure prediction. Using base model - in practice would use fine-tuned SSP model.

Load model for edge case testing

Load model and tokenizer once for all tests

Get detailed information about the underlying base model. Returns comprehensive information about the model architecture, configurat

Model name for testing

Model name for testing

Model name for testing

This function attempts to repair malformed RNA secondary structure representations by ensuring proper bracket matching. It handles common

Convert one‑hot encoded sequences to indices.

Use PlantRNA-FM for plant-specific tasks

Return probabilities via softmax.

Return multi-label probabilities via sigmoid.

Return probabilities via softmax.

Generate predictions using the LoRA-adapted model through the base model interface. This method provides access to the base model's

Predicts on a single text and returns the ensemble result. This method combines predictions from all predictors in the ensemble

Alias for inference method for API compatibility.

This method provides a high-level interface for generating predictions from the pipeline's model. It handles preprocessing and postpr

Return probabilities for each label. This calls the internal convenience routine to accept either raw sequences or already-tokenized

Return probabilities for each label (alias of forward logits).

Return probabilities for each label.

Return probabilities for each label.

Return probabilities for each label.

Return task-appropriate probabilities from head postprocess.

Generate predictions for masked language modeling. Args: sequence_or_inputs: Input sequences or pre-processed inputs

Generate predictions for sequence-level regression. Args: sequence_or_inputs: Input sequences or pre-processed inputs

Generate predictions for matrix regression. Args: sequence_or_inputs: Input sequences or pre-processed inputs

Generate predictions for matrix classification. Args: sequence_or_inputs: Input sequences or pre-processed inputs

This method takes raw sequences or tokenized inputs and returns sequence-level predictions. It processes the inputs through the model

This method takes raw sequences or tokenized inputs and returns multi-label predictions. It applies a threshold to determine

Predict structure and generate comparison visualization

Predicts structures for multiple sequences using multithreading.

Prepare a single data instance for token classification. This method handles both string sequences and dictionary instances

Prepare a single data instance for sequence classification. This method handles both string sequences and dictionary instances

Prepare a single data instance for token regression. This method handles both string sequences and dictionary instances cont

Prepare a single data instance for sequence regression. This method handles both string sequences and dictionary instances c

Prepare a single data instance for multi-label classification. This method handles both string sequences and dictionary instances

Prepare a single training instance with BPP features. This method overrides the parent class to add BPP matrix computation

Prepare input with BPP features. This method: 1. Extracts sequence and label from instance 2. Pads/truncates

This function prints the arguments from a configuration object to the console or a logger. It's useful for debugging and logging experiment p

Print statistics about the multi-label distribution in the dataset. This includes the number of positive labels per sample and per la

Push a model to the hub. This method is not yet implemented and will raise a NotImplementedError. Args: model:

This method is intended to upload custom pipelines to the OmniGenome hub for sharing and distribution. Currently not implemented.

Modify test collection

Configure pytest with custom markers

This function retrieves benchmark information from the OmniGenome hub, either from a remote repository or from a local cache. It supports

This function retrieves pipeline information from the OmniGenome hub, either from a remote repository or from a local cache. It supports

This function is a convenience wrapper for registering the BenchCommand with the argument parser. Args: subparsers: The subparse

This function is a convenience wrapper for registering the RNADesignCommand with the argument parser. Args: subparsers: The subp

This abstract method must be implemented by all subclasses to define their specific command-line interface, including arguments, help

Load RNA design model once for all tests. Using smaller model for faster testing.

Execute the autobench command.

Execute the autoinfer command.

Execute the autotrain command.

This function sets up logging, constructs the command to execute (potentially with `accelerate launch`), and runs it.

Execute the RNA design command.