r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

custom progress bar Example: >>> pbar = ProgressBar(n_total=30,desc='training') >>> step = 2 >>> pbar(step=step)

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Flo

A single training/test example for simple sequence classification.

r""" Compute a single vector summary of a sequence hidden states according to various possibilities: Args of the config class: sum

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

Multi-class Focal loss implementation

r"""A placeholder identity operator that is argument-insensitive.

用于评价模型，计算每个标签的精确率，召回率，F1分数

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

Implements Adam algorithm with weight decay fix. Parameters: lr (float): learning rate. Default 1e-3. betas (tuple of 2 floa

Conditional random field. This module implements a conditional random field [LMP01]_. The forward computation of this class computes the log l

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

Runs basic tokenization (punctuation splitting, lower casing, etc.).

Example # 初始化 fgm = FGM(model,epsilon=1,emb_name='word_embeddings.') for batch_input, batch_label in data: # 正常训练

Writes InputFeature to TF example file.

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

A single training/test example for token classification.

A single training/test example for token classification.

A single set of features of data.

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

Fake example so the num input examples is a multiple of the batch size. When running eval/predict on the TPU, we need to pad the number of examples

Compute SQuAD 2.0 answer class from classification and start tokens hidden states.

Compute SQuAD end_logits from sequence hidden states and start token hidden state.

Compute SQuAD start_logits from sequence hidden states.

A basic Adam optimizer that includes "correct" L2 weight decay.

A basic Adam optimizer that includes "correct" L2 weight decay.

Construct the embeddings from word, position and token_type embeddings.

Construct the embeddings from word, position and token_type embeddings.

r""" **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``: Labels for computing the

Runs basic tokenization (punctuation splitting, lower casing, etc.).

Runs basic tokenization (punctuation splitting, lower casing, etc.).

Configuration for `BertModel`.

Construct the embeddings from word, position and token_type embeddings.

r""" Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs: **last_hidden_state**: ``torch.Float

A wrapper for iterating `torch.utils.data.DataLoader` with the ability to reset itself while `StopIteration` is raised.

Exponentially increases the learning rate between two boundaries over a number of iterations. Arguments: optimizer (torch.optim.Optim