Loads a data file into a list of `InputBatch`s `cls_token_at_end` define the location of the CLS token: - False (Default, BERT/XL

计算每种标签预测正确的个数(对应精确率、召回率计算公式上的tp)，用于后面精确率以及召回率的计算

return a single dataloader for prediction

Find the most likely tag sequence using Viterbi algorithm. Args: emissions (`~torch.Tensor`): Emission score tensor of size

使用维特比算法对给定观测序列求状态序列， 这里就是对字组成的序列,求其对应的标注。 维特比算法实际是用动态规划解隐马尔可夫模型预测问题，即用动态规划求概率最大路径（最优路径） 这时一条路径对应着一个状态序列

See base class.

Gets entities from sequence. note: BIO Args: seq (list): sequence of labels. Returns: list: list of (chunk_type, chunk_sta

Gets entities from sequence. note: BIOS Args: seq (list): sequence of labels. Returns: list: list of (chunk_type, chunk_st

Create a schedule with a learning rate that decreases linearly after linearly increasing during a warmup period.

See base class.

See base class.

训练并评估hmm模型

将targets中的数转化为在[T*T]大小序列中的索引,T是标注的种类

Example: >>> init_logger(log_file) >>> logger.info("abc'")

: yield an index and a slice, that from the day

Load the latest checkpoint.

加载json文件 :param json_path: :param file_name: :return:

deliver arguments from json file to program :param param_dict: a dict that contains model info :return: no return

This function prints and plots the confusion matrix. Normalization can be applied by setting `normalize=True`.

setup GPU device if available, move model into configured device # 如果n_gpu_use为数字，则使用range生成list # 如果输入的是一个list，则默认使用list[0]作为controller

Preprocessing of adjacency matrix for simple GCN model and conversion to tuple representation.

Method to reset internal variables.

Initialize the transition parameters. The parameters will be initialized randomly from a uniform distribution between -0.1 and 0.1.