Serializes this instance to a Python dictionary.

Serializes this instance to a Python dictionary.

Multi-headed, multi-layer Transformer from "Attention is All You Need". This is almost an exact implementation of the original Transformer encoder.

Multi-headed, multi-layer Transformer from "Attention is All You Need". This is almost an exact implementation of the original Transformer encoder.

Constructs BertConfig. Args: vocab_size: Vocabulary size of `inputs_ids` in `BertModel`. hidden_size: Size of the encoder layers an

Constructor for BertModel. Args: config: `BertConfig` instance. is_training: bool. rue for training model, false for eval model. Co

Constructs a BasicTokenizer. Args: do_lower_case: Whether to lower case the input.

Constructs a AdamWeightDecayOptimizer.

Constructs BertConfig. Args: vocab_size: Vocabulary size of `inputs_ids` in `BertModel`. hidden_size: Size of the encoder layers an

Constructor for BertModel. Args: config: `BertConfig` instance. is_training: bool. rue for training model, false for eval model. Co

input_emb: [batch_size, nstep, hidden_size] extra_emb: [batch_size, hidden_size]

得到一个3D的mask [a, p, n], 对应triplet（a, p, n）是valid的位置是True ---------------------------------- Args: labels: 对应训练数据的l

:param embeddings: [batch_size, steps, embedding_dim] :return:

计算两两embedding的距离 ------------------------------------------ Args： embedding: 特征向量， 大小（batch_size, vector_size）

Truncates a sequence pair in place to the maximum length.

triplet loss of a batch ------------------------------- Args: labels: 标签数据，shape = （batch_size,）

Batch normalization on convolutional maps. Ref.: http://stackoverflow.com/questions/33949786/how-could-i-use-batch-normalization-in-tensorflo

Converts `text` to Unicode (if it's not already), assuming utf-8 input.

Creates an optimizer training op.

Constructs a `BertConfig` from a json file of parameters.

Gaussian Error Linear Unit. This is a smoother version of the RELU. Original paper: https://arxiv.org/abs/1606.08415 Args: input_tensor:

Gaussian Error Linear Unit. This is a smoother version of the RELU. Original paper: https://arxiv.org/abs/1606.08415 Args: input_tensor:

Compute the union of the current variables and checkpoint variables.

Compute the union of the current variables and checkpoint variables.

gen datasets, convert word into word ids. :param file_path: :return: [[query, pos sample, 4 neg sample]], shape = [n, 6]

gen datasets, convert word into word ids. :param file_path: :return: [[query, prefix, label]], shape = [n, 3]

gen datasets, convert word into word ids. :param file_path: :return: [[query, pos sample, 4 neg sample]], shape = [n, 6]

Gets output of the embedding lookup (i.e., input to the transformer). Returns: float Tensor of shape [batch_size, seq_length, hidden_size]

Gets output of the embedding lookup (i.e., input to the transformer). Returns: float Tensor of shape [batch_size, seq_length, hidden_size]

使用LCQMC数据集，并将每个query其转为word_id，

Gets final hidden layer of encoder. Returns: float Tensor of shape [batch_size, seq_length, hidden_size] corresponding to the final

From the paper: "... employed label smoothing of epsilon = 0.1. This hurts perplexity, as the model learns to be more unsure, but imp

Returns text encoded in a way suitable for print or `tf.logging`.

Serializes this instance to a JSON string.

Serializes this instance to a JSON string.

Tokenizes a piece of text.

Tokenizes a piece of text into its word pieces. This uses a greedy longest-match-first algorithm to perform tokenization using the