Linearized nodes with varids replaced by instances

Get the values of nonempty fields as a list.

Get the values of nonempty fields as a list.

Callback before the training starts

adopted from Timothy Dozat https://github.com/tdozat/Parser/blob/master/lib/linalg.py Args: output_size: input_size: Returns:

The index of ``PAD`` token.

Compute the performance metrics: TP, FP, FN, TN Args: y_true: 2d array. Ground truth (correct) target values. y_pred: 2d array. Estim

* Computes normalized vector mse loss at position 0 along `length` dimension. This is the loss used in BERT-PKD, see `Patient Knowledge Distillat

powerset([1,2,3]) --> () (1,) (2,) (3,) (1,2) (1,3) (2,3) (1,2,3) Args: iterable: Returns:

Classify text. Args: text: A document or a list of documents. topk: ``True`` or ``int`` to return the top-k

Classify text. Args: text: A document or a list of documents. topk: ``True`` or ``int`` to return the top-k

Classify text. Args: text: A document or a list of documents. topk: ``True`` or ``int`` to return the top-k

Predict the similarity between sentence pairs. Args: data: Sentence pairs. batch_size: The number of samples in a ba

Sentence split. Args: data: A paragraph or a list of paragraphs. batch_size: Number of samples per batch.

Predict on data. Args: data: A sentence or a list of sentences. tasks: The tasks to predict. skip_tasks:

Predict on data fed by user. Users shall avoid directly call this method since it is not guarded with ``torch.no_grad`` and will introduces un

``True`` if this tree is projective.

Fix the relation prediction by heuristic rules Args: rel_probs(NDArray): seq_len x rel_size length: real sentence length ensure

Fix the relation prediction by heuristic rules Args: rel_probs(NDArray): seq_len x rel_size length: real sentence length ensure