Code
stringlengths
103
85.9k
Summary
listlengths
0
94
Please provide a description of the function:def train(args): train_file = args.input test_file = args.validation ngram_range = args.ngrams logging.info('Ngrams range for the training run : %s', ngram_range) logging.info('Loading Training data') train_labels, train_data = read_input_data(tr...
[ "Training function that orchestrates the Classification! " ]
Please provide a description of the function:def hybrid_forward(self, F, inputs, token_types, valid_length=None): # pylint: disable=arguments-differ # pylint: disable=unused-argument bert_output = self.bert(inputs, token_types, valid_length) output = self.span_classifier(bert_ou...
[ "Generate the unnormalized score for the given the input sequences.\n\n Parameters\n ----------\n inputs : NDArray, shape (batch_size, seq_length)\n Input words for the sequences.\n token_types : NDArray, shape (batch_size, seq_length)\n Token types for the sequence...
Please provide a description of the function:def hybrid_forward(self, F, pred, label): # pylint: disable=arguments-differ pred = F.split(pred, axis=2, num_outputs=2) start_pred = pred[0].reshape((0, -3)) start_label = label[0] end_pred = pred[1].reshape((0, -3)) end_lab...
[ "\n Parameters\n ----------\n pred : NDArray, shape (batch_size, seq_length, 2)\n BERTSquad forward output.\n label : list, length is 2, each shape is (batch_size,1)\n label[0] is the starting position of the answer,\n label[1] is the ending position of t...
Please provide a description of the function:def encode(self, inputs, states=None, valid_length=None): return self.encoder(self.src_embed(inputs), states, valid_length)
[ "Encode the input sequence.\n\n Parameters\n ----------\n inputs : NDArray\n states : list of NDArrays or None, default None\n valid_length : NDArray or None, default None\n\n Returns\n -------\n outputs : list\n Outputs of the encoder.\n " ]
Please provide a description of the function:def decode_seq(self, inputs, states, valid_length=None): outputs, states, additional_outputs =\ self.decoder.decode_seq(inputs=self.tgt_embed(inputs), states=states, valid_le...
[ "Decode given the input sequence.\n\n Parameters\n ----------\n inputs : NDArray\n states : list of NDArrays\n valid_length : NDArray or None, default None\n\n Returns\n -------\n output : NDArray\n The output of the decoder. Shape is (batch_size, l...
Please provide a description of the function:def decode_step(self, step_input, states): step_output, states, step_additional_outputs =\ self.decoder(self.tgt_embed(step_input), states) step_output = self.tgt_proj(step_output) return step_output, states, step_additional_outpu...
[ "One step decoding of the translation model.\n\n Parameters\n ----------\n step_input : NDArray\n Shape (batch_size,)\n states : list of NDArrays\n\n Returns\n -------\n step_output : NDArray\n Shape (batch_size, C_out)\n states : list\n ...
Please provide a description of the function:def forward(self, src_seq, tgt_seq, src_valid_length=None, tgt_valid_length=None): #pylint: disable=arguments-differ additional_outputs = [] encoder_outputs, encoder_additional_outputs = self.encode(src_seq, ...
[ "Generate the prediction given the src_seq and tgt_seq.\n\n This is used in training an NMT model.\n\n Parameters\n ----------\n src_seq : NDArray\n tgt_seq : NDArray\n src_valid_length : NDArray or None\n tgt_valid_length : NDArray or None\n\n Returns\n ...
Please provide a description of the function:def create_subword_function(subword_function_name, **kwargs): create_ = registry.get_create_func(SubwordFunction, 'token embedding') return create_(subword_function_name, **kwargs)
[ "Creates an instance of a subword function." ]
Please provide a description of the function:def _index_special_tokens(self, unknown_token, special_tokens): self._idx_to_token = [unknown_token] if unknown_token else [] if not special_tokens: self._reserved_tokens = None else: self._reserved_tokens = special_t...
[ "Indexes unknown and reserved tokens." ]
Please provide a description of the function:def _index_counter_keys(self, counter, unknown_token, special_tokens, max_size, min_freq): unknown_and_special_tokens = set(special_tokens) if special_tokens else set() if unknown_token: unknown_and_special_t...
[ "Indexes keys of `counter`.\n\n\n Indexes keys of `counter` according to frequency thresholds such as `max_size` and\n `min_freq`.\n " ]
Please provide a description of the function:def set_embedding(self, *embeddings): if len(embeddings) == 1 and embeddings[0] is None: self._embedding = None return for embs in embeddings: assert isinstance(embs, emb.TokenEmbedding), \ 'The a...
[ "Attaches one or more embeddings to the indexed text tokens.\n\n\n Parameters\n ----------\n embeddings : None or tuple of :class:`gluonnlp.embedding.TokenEmbedding` instances\n The embedding to be attached to the indexed tokens. If a tuple of multiple embeddings\n are pro...
Please provide a description of the function:def to_tokens(self, indices): to_reduce = False if not isinstance(indices, (list, tuple)): indices = [indices] to_reduce = True max_idx = len(self._idx_to_token) - 1 tokens = [] for idx in indices: ...
[ "Converts token indices to tokens according to the vocabulary.\n\n\n Parameters\n ----------\n indices : int or list of ints\n A source token index or token indices to be converted.\n\n\n Returns\n -------\n str or list of strs\n A token or a list of t...
Please provide a description of the function:def to_json(self): if self._embedding: warnings.warn('Serialization of attached embedding ' 'to json is not supported. ' 'You may serialize the embedding to a binary format ' ...
[ "Serialize Vocab object to json string.\n\n This method does not serialize the underlying embedding.\n " ]
Please provide a description of the function:def from_json(cls, json_str): vocab_dict = json.loads(json_str) unknown_token = vocab_dict.get('unknown_token') vocab = cls(unknown_token=unknown_token) vocab._idx_to_token = vocab_dict.get('idx_to_token') vocab._token_to_idx...
[ "Deserialize Vocab object from json string.\n\n Parameters\n ----------\n json_str : str\n Serialized json string of a Vocab object.\n\n\n Returns\n -------\n Vocab\n " ]
Please provide a description of the function:def train(data_train, model, nsp_loss, mlm_loss, vocab_size, ctx): hvd.broadcast_parameters(model.collect_params(), root_rank=0) mlm_metric = nlp.metric.MaskedAccuracy() nsp_metric = nlp.metric.MaskedAccuracy() mlm_metric.reset() nsp_metric.reset() ...
[ "Training function." ]
Please provide a description of the function:def train(): log.info('Loader Train data...') if version_2: train_data = SQuAD('train', version='2.0') else: train_data = SQuAD('train', version='1.1') log.info('Number of records in Train data:{}'.format(len(train_data))) train_data...
[ "Training function.", "set new learning rate" ]
Please provide a description of the function:def evaluate(): log.info('Loader dev data...') if version_2: dev_data = SQuAD('dev', version='2.0') else: dev_data = SQuAD('dev', version='1.1') log.info('Number of records in Train data:{}'.format(len(dev_data))) dev_dataset = dev_d...
[ "Evaluate the model on validation dataset.\n " ]
Please provide a description of the function:def _pad_arrs_to_max_length(arrs, pad_axis, pad_val, use_shared_mem, dtype): if isinstance(arrs[0], mx.nd.NDArray): dtype = arrs[0].dtype if dtype is None else dtype arrs = [arr.asnumpy() for arr in arrs] elif not isinstance(arrs[0], np.ndarray):...
[ "Inner Implementation of the Pad batchify\n\n Parameters\n ----------\n arrs : list\n pad_axis : int\n pad_val : number\n use_shared_mem : bool, default False\n\n Returns\n -------\n ret : NDArray\n original_length : NDArray\n " ]
Please provide a description of the function:def train(self, train_file, dev_file, test_file, save_dir, pretrained_embeddings=None, min_occur_count=2, lstm_layers=3, word_dims=100, tag_dims=100, dropout_emb=0.33, lstm_hiddens=400, dropout_lstm_input=0.33, dropout_lstm_hidden=0.33, mlp_arc_si...
[ "Train a deep biaffine dependency parser\n\n Parameters\n ----------\n train_file : str\n path to training set\n dev_file : str\n path to dev set\n test_file : str\n path to test set\n save_dir : str\n a directory for saving model...
Please provide a description of the function:def load(self, path): config = _Config.load(os.path.join(path, 'config.pkl')) config.save_dir = path # redirect root path to what user specified self._vocab = vocab = ParserVocabulary.load(config.save_vocab_path) with mx.Context(mxne...
[ "Load from disk\n\n Parameters\n ----------\n path : str\n path to the directory which typically contains a config.pkl file and a model.bin file\n\n Returns\n -------\n DepParser\n parser itself\n " ]
Please provide a description of the function:def evaluate(self, test_file, save_dir=None, logger=None, num_buckets_test=10, test_batch_size=5000): parser = self._parser vocab = self._vocab with mx.Context(mxnet_prefer_gpu()): UAS, LAS, speed = evaluate_official_script(parser...
[ "Run evaluation on test set\n\n Parameters\n ----------\n test_file : str\n path to test set\n save_dir : str\n where to store intermediate results and log\n logger : logging.logger\n logger for printing results\n num_buckets_test : int\n ...
Please provide a description of the function:def parse(self, sentence): words = np.zeros((len(sentence) + 1, 1), np.int32) tags = np.zeros((len(sentence) + 1, 1), np.int32) words[0, 0] = ParserVocabulary.ROOT tags[0, 0] = ParserVocabulary.ROOT vocab = self._vocab ...
[ "Parse raw sentence into ConllSentence\n\n Parameters\n ----------\n sentence : list\n a list of (word, tag) tuples\n\n Returns\n -------\n ConllSentence\n ConllSentence object\n " ]
Please provide a description of the function:def apply_weight_drop(block, local_param_regex, rate, axes=(), weight_dropout_mode='training'): if not rate: return existing_params = _find_params(block, local_param_regex) for (local_param_name, param), \ (ref_para...
[ "Apply weight drop to the parameter of a block.\n\n Parameters\n ----------\n block : Block or HybridBlock\n The block whose parameter is to be applied weight-drop.\n local_param_regex : str\n The regex for parameter names used in the self.params.get(), such as 'weight'.\n rate : float\...
Please provide a description of the function:def _get_rnn_cell(mode, num_layers, input_size, hidden_size, dropout, weight_dropout, var_drop_in, var_drop_state, var_drop_out, skip_connection, proj_size=None, cell_clip=None, proj_clip=None): assert mode == '...
[ "create rnn cell given specs\n\n Parameters\n ----------\n mode : str\n The type of RNN cell to use. Options are 'lstmpc', 'rnn_tanh', 'rnn_relu', 'lstm', 'gru'.\n num_layers : int\n The number of RNN cells in the encoder.\n input_size : int\n The initial input size of in the RNN...
Please provide a description of the function:def _get_rnn_layer(mode, num_layers, input_size, hidden_size, dropout, weight_dropout): if mode == 'rnn_relu': rnn_block = functools.partial(rnn.RNN, activation='relu') elif mode == 'rnn_tanh': rnn_block = functools.partial(rnn.RNN, activation='t...
[ "create rnn layer given specs" ]
Please provide a description of the function:def hybrid_forward(self, F, x, sampled_values, label, w_all, b_all): sampled_candidates, expected_count_sampled, expected_count_true = sampled_values # (num_sampled, in_unit) w_sampled = w_all.slice(begin=(0, 0), end=(self._num_sampled, None)...
[ "Forward computation." ]
Please provide a description of the function:def hybrid_forward(self, F, x, sampled_values, label, weight, bias): sampled_candidates, _, _ = sampled_values # (batch_size,) label = F.reshape(label, shape=(-1,)) # (num_sampled+batch_size,) ids = F.concat(sampled_candidates...
[ "Forward computation." ]
Please provide a description of the function:def forward(self, x, sampled_values, label): sampled_candidates, _, _ = sampled_values # (batch_size,) label = label.reshape(shape=(-1,)) # (num_sampled+batch_size,) ids = nd.concat(sampled_candidates, label, dim=0) # ...
[ "Forward computation." ]
Please provide a description of the function:def _extract_and_flatten_nested_structure(data, flattened=None): if flattened is None: flattened = [] structure = _extract_and_flatten_nested_structure(data, flattened) return structure, flattened if isinstance(data, list): return...
[ "Flatten the structure of a nested container to a list.\n\n Parameters\n ----------\n data : A single NDArray/Symbol or nested container with NDArrays/Symbol.\n The nested container to be flattened.\n flattened : list or None\n The container thats holds flattened result.\n Returns\n ...
Please provide a description of the function:def _reconstruct_flattened_structure(structure, flattened): if isinstance(structure, list): return list(_reconstruct_flattened_structure(x, flattened) for x in structure) elif isinstance(structure, tuple): return tuple(_reconstruct_flattened_stru...
[ "Reconstruct the flattened list back to (possibly) nested structure.\n\n Parameters\n ----------\n structure : An integer or a nested container with integers.\n The extracted structure of the container of `data`.\n flattened : list or None\n The container thats holds flattened result.\n ...
Please provide a description of the function:def _expand_to_beam_size(data, beam_size, batch_size, state_info=None): assert not state_info or isinstance(state_info, (type(data), dict)), \ 'data and state_info doesn\'t match, ' \ 'got: {} vs {}.'.format(type(state_info), type(data)) ...
[ "Tile all the states to have batch_size * beam_size on the batch axis.\n\n Parameters\n ----------\n data : A single NDArray/Symbol or nested container with NDArrays/Symbol\n Each NDArray/Symbol should have shape (N, ...) when state_info is None,\n or same as the layout in state_info when it'...
Please provide a description of the function:def hybrid_forward(self, F, samples, valid_length, outputs, scores, beam_alive_mask, states): beam_size = self._beam_size # outputs: (batch_size, beam_size, vocab_size) outputs = outputs.reshape(shape=(-4, -1, beam_size, 0)) smoothed_...
[ "\n Parameters\n ----------\n F\n samples : NDArray or Symbol\n The current samples generated by beam search. Shape (batch_size, beam_size, L)\n valid_length : NDArray or Symbol\n The current valid lengths of the samples\n outputs: NDArray or Symbol\n ...
Please provide a description of the function:def hybrid_forward(self, F, inputs, states): # pylint: disable=arguments-differ batch_size = self._batch_size beam_size = self._beam_size vocab_size = self._vocab_size # Tile the states and inputs to have shape (batch_size * beam_si...
[ "Sample by beam search.\n\n Parameters\n ----------\n F\n inputs : NDArray or Symbol\n The initial input of the decoder. Shape is (batch_size,).\n states : Object that contains NDArrays or Symbols\n The initial states of the decoder.\n Returns\n ...
Please provide a description of the function:def data(self, ctx=None): d = self._check_and_get(self._data, ctx) if self._rate: d = nd.Dropout(d, self._rate, self._mode, self._axes) return d
[ "Returns a copy of this parameter on one context. Must have been\n initialized on this context before.\n\n Parameters\n ----------\n ctx : Context\n Desired context.\n Returns\n -------\n NDArray on ctx\n " ]
Please provide a description of the function:def elmo_2x1024_128_2048cnn_1xhighway(dataset_name=None, pretrained=False, ctx=mx.cpu(), root=os.path.join(get_home_dir(), 'models'), **kwargs): r predefined_args = {'rnn_type': 'lstmpc', 'output_size': 12...
[ "ELMo 2-layer BiLSTM with 1024 hidden units, 128 projection size, 1 highway layer.\n\n Parameters\n ----------\n dataset_name : str or None, default None\n The dataset name on which the pre-trained model is trained.\n Options are 'gbw'.\n pretrained : bool, default False\n Whether t...
Please provide a description of the function:def hybrid_forward(self, F, inputs): # pylint: disable=arguments-differ # the character id embedding # (batch_size * sequence_length, max_chars_per_token, embed_dim) character_embedding = self._char_embedding(inputs.reshape((-1, self....
[ "\n Compute context insensitive token embeddings for ELMo representations.\n\n Parameters\n ----------\n inputs : NDArray\n Shape (batch_size, sequence_length, max_character_per_token)\n of character ids representing the current batch.\n\n Returns\n --...
Please provide a description of the function:def hybrid_forward(self, F, inputs, states=None, mask=None): # pylint: disable=arguments-differ type_representation = self._elmo_char_encoder(inputs) type_representation = type_representation.transpose(axes=(1, 0, 2)) lstm_outputs, s...
[ "\n Parameters\n ----------\n inputs : NDArray\n Shape (batch_size, sequence_length, max_character_per_token)\n of character ids representing the current batch.\n states : (list of list of NDArray, list of list of NDArray)\n The states. First tuple elemen...
Please provide a description of the function:def awd_lstm_lm_1150(dataset_name=None, vocab=None, pretrained=False, ctx=cpu(), root=os.path.join(get_home_dir(), 'models'), **kwargs): r predefined_args = {'embed_size': 400, 'hidden_size': 1150, 'm...
[ "3-layer LSTM language model with weight-drop, variational dropout, and tied weights.\n\n Embedding size is 400, and hidden layer size is 1150.\n\n Parameters\n ----------\n dataset_name : str or None, default None\n The dataset name on which the pre-trained model is trained.\n Options are...
Please provide a description of the function:def standard_lstm_lm_200(dataset_name=None, vocab=None, pretrained=False, ctx=cpu(), root=os.path.join(get_home_dir(), 'models'), **kwargs): r predefined_args = {'embed_size': 200, 'hidden_size': 200, ...
[ "Standard 2-layer LSTM language model with tied embedding and output weights.\n\n Both embedding and hidden dimensions are 200.\n\n Parameters\n ----------\n dataset_name : str or None, default None\n The dataset name on which the pre-trained model is trained.\n Options are 'wikitext-2'. I...
Please provide a description of the function:def big_rnn_lm_2048_512(dataset_name=None, vocab=None, pretrained=False, ctx=cpu(), root=os.path.join(get_home_dir(), 'models'), **kwargs): r predefined_args = {'embed_size': 512, 'hidden_size': 2048, ...
[ "Big 1-layer LSTMP language model.\n\n Both embedding and projection size are 512. Hidden size is 2048.\n\n Parameters\n ----------\n dataset_name : str or None, default None\n The dataset name on which the pre-trained model is trained.\n Options are 'gbw'. If specified, then the returned ...
Please provide a description of the function:def forward(self, inputs, begin_state=None): # pylint: disable=arguments-differ encoded = self.embedding(inputs) if begin_state is None: begin_state = self.begin_state(batch_size=inputs.shape[1]) out_states = [] for i, (e,...
[ "Implement forward computation.\n\n Parameters\n -----------\n inputs : NDArray\n input tensor with shape `(sequence_length, batch_size)`\n when `layout` is \"TNC\".\n begin_state : list\n initial recurrent state tensor with length equals to num_layers.\n...
Please provide a description of the function:def forward(self, inputs, begin_state): # pylint: disable=arguments-differ encoded = self.embedding(inputs) length = inputs.shape[0] batch_size = inputs.shape[1] encoded, state = self.encoder.unroll(length, encoded, begin_state, ...
[ "Implement forward computation.\n\n Parameters\n -----------\n inputs : NDArray\n input tensor with shape `(sequence_length, batch_size)`\n when `layout` is \"TNC\".\n begin_state : list\n initial recurrent state tensor with length equals to num_layers*2....
Please provide a description of the function:def _get_cell_type(cell_type): if isinstance(cell_type, str): if cell_type == 'lstm': return rnn.LSTMCell elif cell_type == 'gru': return rnn.GRUCell elif cell_type == 'relu_rnn': return partial(rnn.RNNCell...
[ "Get the object type of the cell by parsing the input\n\n Parameters\n ----------\n cell_type : str or type\n\n Returns\n -------\n cell_constructor: type\n The constructor of the RNNCell\n " ]
Please provide a description of the function:def _get_context(center_idx, sentence_boundaries, window_size, random_window_size, seed): random.seed(seed + center_idx) sentence_index = np.searchsorted(sentence_boundaries, center_idx) sentence_start, sentence_end = _get_sentence_start_en...
[ "Compute the context with respect to a center word in a sentence.\n\n Takes an numpy array of sentences boundaries.\n\n " ]
Please provide a description of the function:def model(dropout, vocab, model_mode, output_size): textCNN = SentimentNet(dropout=dropout, vocab_size=len(vocab), model_mode=model_mode,\ output_size=output_size) textCNN.hybridize() return textCNN
[ "Construct the model." ]
Please provide a description of the function:def init(textCNN, vocab, model_mode, context, lr): textCNN.initialize(mx.init.Xavier(), ctx=context, force_reinit=True) if model_mode != 'rand': textCNN.embedding.weight.set_data(vocab.embedding.idx_to_vec) if model_mode == 'multichannel': t...
[ "Initialize parameters." ]
Please provide a description of the function:def preprocess_dataset(dataset, transform, num_workers=8): worker_fn = partial(_worker_fn, transform=transform) start = time.time() pool = mp.Pool(num_workers) dataset_transform = [] dataset_len = [] for data in pool.map(worker_fn, dataset): ...
[ "Use multiprocessing to perform transform for dataset.\n\n Parameters\n ----------\n dataset: dataset-like object\n Source dataset.\n transform: callable\n Transformer function.\n num_workers: int, default 8\n The number of multiprocessing workers to use for data preprocessing.\n...
Please provide a description of the function:def get_model(name, dataset_name='wikitext-2', **kwargs): models = {'standard_lstm_lm_200' : standard_lstm_lm_200, 'standard_lstm_lm_650' : standard_lstm_lm_650, 'standard_lstm_lm_1500': standard_lstm_lm_1500, 'awd_lstm_lm_1...
[ "Returns a pre-defined model by name.\n\n Parameters\n ----------\n name : str\n Name of the model.\n dataset_name : str or None, default 'wikitext-2'.\n The dataset name on which the pre-trained model is trained.\n For language model, options are 'wikitext-2'.\n For ELMo, Op...
Please provide a description of the function:def _masked_softmax(F, att_score, mask, dtype): if mask is not None: # Fill in the masked scores with a very small value neg = -1e4 if np.dtype(dtype) == np.float16 else -1e18 att_score = F.where(mask, att_score, neg * F.ones_like(att_score))...
[ "Ignore the masked elements when calculating the softmax\n\n Parameters\n ----------\n F : symbol or ndarray\n att_score : Symborl or NDArray\n Shape (batch_size, query_length, memory_length)\n mask : Symbol or NDArray or None\n Shape (batch_size, query_length, memory_length)\n Retur...
Please provide a description of the function:def _read_by_weight(self, F, att_weights, value): output = F.batch_dot(att_weights, value) return output
[ "Read from the value matrix given the attention weights.\n\n Parameters\n ----------\n F : symbol or ndarray\n att_weights : Symbol or NDArray\n Attention weights.\n For single-head attention,\n Shape (batch_size, query_length, memory_length).\n ...
Please provide a description of the function:def translate(self, src_seq, src_valid_length): batch_size = src_seq.shape[0] encoder_outputs, _ = self._model.encode(src_seq, valid_length=src_valid_length) decoder_states = self._model.decoder.init_state_from_encoder(encoder_outputs, ...
[ "Get the translation result given the input sentence.\n\n Parameters\n ----------\n src_seq : mx.nd.NDArray\n Shape (batch_size, length)\n src_valid_length : mx.nd.NDArray\n Shape (batch_size,)\n\n Returns\n -------\n samples : NDArray\n ...
Please provide a description of the function:def evaluate_official_script(parser, vocab, num_buckets_test, test_batch_size, test_file, output_file, debug=False): if output_file is None: output_file = tempfile.NamedTemporaryFile().name data_loader = DataLoader(test_file,...
[ "Evaluate parser on a data set\n\n Parameters\n ----------\n parser : BiaffineParser\n biaffine parser\n vocab : ParserVocabulary\n vocabulary built from data set\n num_buckets_test : int\n size of buckets (cluster sentences into this number of clusters)\n test_batch_size : in...
Please provide a description of the function:def parameter_from_numpy(self, name, array): p = self.params.get(name, shape=array.shape, init=mx.init.Constant(array)) return p
[ " Create parameter with its value initialized according to a numpy tensor\n\n Parameters\n ----------\n name : str\n parameter name\n array : np.ndarray\n initiation value\n\n Returns\n -------\n mxnet.gluon.parameter\n a parameter ob...
Please provide a description of the function:def parameter_init(self, name, shape, init): p = self.params.get(name, shape=shape, init=init) return p
[ "Create parameter given name, shape and initiator\n\n Parameters\n ----------\n name : str\n parameter name\n shape : tuple\n parameter shape\n init : mxnet.initializer\n an initializer\n\n Returns\n -------\n mxnet.gluon.param...
Please provide a description of the function:def forward(self, word_inputs, tag_inputs, arc_targets=None, rel_targets=None): is_train = autograd.is_training() def flatten_numpy(ndarray): return np.reshape(ndarray, (-1,), 'F') batch_size = word_inputs.shape[1] ...
[ "Run decoding\n\n Parameters\n ----------\n word_inputs : mxnet.ndarray.NDArray\n word indices of seq_len x batch_size\n tag_inputs : mxnet.ndarray.NDArray\n tag indices of seq_len x batch_size\n arc_targets : mxnet.ndarray.NDArray\n gold arc indic...
Please provide a description of the function:def save_parameters(self, filename): params = self._collect_params_with_prefix() if self.pret_word_embs: # don't save word embeddings inside model params.pop('pret_word_embs.weight', None) arg_dict = {key: val._reduce() for key, ...
[ "Save model\n\n Parameters\n ----------\n filename : str\n path to model file\n " ]
Please provide a description of the function:def _worker_fn(samples, batchify_fn, dataset=None): # pylint: disable=unused-argument # it is required that each worker process has to fork a new MXIndexedRecordIO handle # preserving dataset as global variable can save tons of overhead and is safe in new pr...
[ "Function for processing data in worker process." ]
Please provide a description of the function:def _thread_worker_fn(samples, batchify_fn, dataset): if isinstance(samples[0], (list, tuple)): batch = [batchify_fn([dataset[i] for i in shard]) for shard in samples] else: batch = batchify_fn([dataset[i] for i in samples]) return batch
[ "Threadpool worker function for processing data." ]
Please provide a description of the function:def create(embedding_name, **kwargs): create_text_embedding = registry.get_create_func(TokenEmbedding, 'token embedding') return create_text_embedding(embedding_name, **kwargs)
[ "Creates an instance of token embedding.\n\n\n Creates a token embedding instance by loading embedding vectors from an externally hosted\n pre-trained token embedding file, such as those of GloVe and FastText. To get all the valid\n `embedding_name` and `source`, use :func:`gluonnlp.embedding.list_sources`...
Please provide a description of the function:def list_sources(embedding_name=None): text_embedding_reg = registry.get_registry(TokenEmbedding) if embedding_name is not None: embedding_name = embedding_name.lower() if embedding_name not in text_embedding_reg: raise KeyError('Ca...
[ "Get valid token embedding names and their pre-trained file names.\n\n\n To load token embedding vectors from an externally hosted pre-trained token embedding file,\n such as those of GloVe and FastText, one should use\n `gluonnlp.embedding.create(embedding_name, source)`. This method returns all the\n ...
Please provide a description of the function:def _load_embedding(self, pretrained_file_path, elem_delim, encoding='utf8'): pretrained_file_path = os.path.expanduser(pretrained_file_path) if not os.path.isfile(pretrained_file_path): raise ValueError('`pretra...
[ "Load embedding vectors from a pre-trained token embedding file.\n\n Both text files and TokenEmbedding serialization files are supported.\n elem_delim and encoding are ignored for non-text files.\n\n For every unknown token, if its representation `self.unknown_token` is encountered in the\n ...
Please provide a description of the function:def _load_embedding_txt(self, pretrained_file_path, elem_delim, encoding='utf8'): vec_len = None all_elems = [] tokens = set() loaded_unknown_vec = None with io.open(pretrained_file_path, 'rb') as f: for line_num,...
[ "Load embedding vectors from a pre-trained token embedding file.\n\n For every unknown token, if its representation `self.unknown_token` is encountered in the\n pre-trained token embedding file, index 0 of `self.idx_to_vec` maps to the pre-trained token\n embedding vector loaded from the file; ...
Please provide a description of the function:def _load_embedding_serialized(self, pretrained_file_path): deserialized_embedding = TokenEmbedding.deserialize(pretrained_file_path) if deserialized_embedding.unknown_token: # Some .npz files on S3 may contain an unknown token and its ...
[ "Load embedding vectors from a pre-trained token embedding file.\n\n For every unknown token, if its representation `self.unknown_token` is encountered in the\n pre-trained token embedding file, index 0 of `self.idx_to_vec` maps to the pre-trained token\n embedding vector loaded from the file; ...
Please provide a description of the function:def _check_vector_update(self, tokens, new_embedding): assert self._idx_to_vec is not None, '`idx_to_vec` has not been initialized.' if not isinstance(tokens, (list, tuple)) or len(tokens) == 1: assert isinstance(new_embedding, nd.NDArra...
[ "Check that tokens and embedding are in the format for __setitem__." ]
Please provide a description of the function:def _check_source(cls, source_file_hash, source): embedding_name = cls.__name__.lower() if source not in source_file_hash: raise KeyError('Cannot find pre-trained source {} for token embedding {}. ' 'Valid pre-t...
[ "Checks if a pre-trained token embedding source name is valid.\n\n\n Parameters\n ----------\n source : str\n The pre-trained token embedding source.\n " ]
Please provide a description of the function:def from_file(file_path, elem_delim=' ', encoding='utf8', **kwargs): embedding = TokenEmbedding(**kwargs) embedding._load_embedding(file_path, elem_delim=elem_delim, encoding=encoding) return embedding
[ "Creates a user-defined token embedding from a pre-trained embedding file.\n\n\n This is to load embedding vectors from a user-defined pre-trained token embedding file.\n For example, if `elem_delim` = ' ', the expected format of a custom pre-trained token\n embedding file may look like:\n\n ...
Please provide a description of the function:def serialize(self, file_path, compress=True): if self.unknown_lookup is not None: warnings.warn( 'Serialization of `unknown_lookup` is not supported. ' 'Save it manually and pass the loaded lookup object ' ...
[ "Serializes the TokenEmbedding to a file specified by file_path.\n\n TokenEmbedding is serialized by converting the list of tokens, the\n array of word embeddings and other metadata to numpy arrays, saving all\n in a single (optionally compressed) Zipfile. See\n https://docs.scipy.org/do...
Please provide a description of the function:def deserialize(cls, file_path, **kwargs): # idx_to_token is of dtype 'O' so we need to allow pickle npz_dict = np.load(file_path, allow_pickle=True) unknown_token = npz_dict['unknown_token'] if not unknown_token: unknown...
[ "Create a new TokenEmbedding from a serialized one.\n\n TokenEmbedding is serialized by converting the list of tokens, the\n array of word embeddings and other metadata to numpy arrays, saving all\n in a single (optionally compressed) Zipfile. See\n https://docs.scipy.org/doc/numpy-1.14....
Please provide a description of the function:def evaluate(data_source): log.info('Start predict') tic = time.time() for batch in data_source: inputs, token_types, valid_length = batch out = net(inputs.astype('float32').as_in_context(ctx), token_types.astype('float32')....
[ "Evaluate the model on a mini-batch.\n " ]
Please provide a description of the function:def register(class_=None, **kwargs): def _real_register(class_): # Assert that the passed kwargs are meaningful for kwarg_name, values in kwargs.items(): try: real_args = inspect.getfullargspec(class_).args ex...
[ "Registers a dataset with segment specific hyperparameters.\n\n When passing keyword arguments to `register`, they are checked to be valid\n keyword arguments for the registered Dataset class constructor and are\n saved in the registry. Registered keyword arguments can be retrieved with\n the `list_data...
Please provide a description of the function:def create(name, **kwargs): create_ = registry.get_create_func(Dataset, 'dataset') return create_(name, **kwargs)
[ "Creates an instance of a registered dataset.\n\n Parameters\n ----------\n name : str\n The dataset name (case-insensitive).\n\n Returns\n -------\n An instance of :class:`mxnet.gluon.data.Dataset` constructed with the\n keyword arguments passed to the create function.\n\n " ]
Please provide a description of the function:def list_datasets(name=None): reg = registry.get_registry(Dataset) if name is not None: class_ = reg[name.lower()] return _REGSITRY_NAME_KWARGS[class_] else: return { dataset_name: _REGSITRY_NAME_KWARGS[class_] ...
[ "Get valid datasets and registered parameters.\n\n Parameters\n ----------\n name : str or None, default None\n Return names and registered parameters of registered datasets. If name\n is specified, only registered parameters of the respective dataset are\n returned.\n\n Returns\n ...
Please provide a description of the function:def parse_args(): parser = argparse.ArgumentParser( description='Vocabulary extractor.', formatter_class=argparse.ArgumentDefaultsHelpFormatter) parser.add_argument('--max-size', type=int, default=None) parser.add_argument('--min-freq', type=...
[ "Parse command line arguments." ]
Please provide a description of the function:def get_vocab(args): counter = nlp.data.Counter() start = time.time() for filename in args.files: print('Starting processing of {} after {:.1f} seconds.'.format( filename, time.time() - start)) with open(filename, 'r')...
[ "Compute the vocabulary." ]
Please provide a description of the function:def forward(self, inputs, token_types, valid_length=None): # pylint: disable=arguments-differ _, pooler_out = self.bert(inputs, token_types, valid_length) return self.classifier(pooler_out)
[ "Generate the unnormalized score for the given the input sequences.\n\n Parameters\n ----------\n inputs : NDArray, shape (batch_size, seq_length)\n Input words for the sequences.\n token_types : NDArray, shape (batch_size, seq_length)\n Token types for the sequence...
Please provide a description of the function:def add_parameters(parser): group = parser.add_argument_group('Evaluation arguments') group.add_argument('--eval-batch-size', type=int, default=1024) # Datasets group.add_argument( '--similarity-datasets', type=str, default=nlp.data.wor...
[ "Add evaluation specific parameters to parser." ]
Please provide a description of the function:def validate_args(args): # Check correctness of similarity dataset names for dataset_name in args.similarity_datasets: if dataset_name.lower() not in map( str.lower, nlp.data.word_embedding_evaluation.word_similarity_datas...
[ "Validate provided arguments and act on --help." ]
Please provide a description of the function:def iterate_similarity_datasets(args): for dataset_name in args.similarity_datasets: parameters = nlp.data.list_datasets(dataset_name) for key_values in itertools.product(*parameters.values()): kwargs = dict(zip(parameters.keys(), key_val...
[ "Generator over all similarity evaluation datasets.\n\n Iterates over dataset names, keyword arguments for their creation and the\n created dataset.\n\n " ]
Please provide a description of the function:def iterate_analogy_datasets(args): for dataset_name in args.analogy_datasets: parameters = nlp.data.list_datasets(dataset_name) for key_values in itertools.product(*parameters.values()): kwargs = dict(zip(parameters.keys(), key_values)) ...
[ "Generator over all analogy evaluation datasets.\n\n Iterates over dataset names, keyword arguments for their creation and the\n created dataset.\n\n " ]
Please provide a description of the function:def get_similarity_task_tokens(args): tokens = set() for _, _, dataset in iterate_similarity_datasets(args): tokens.update( itertools.chain.from_iterable((d[0], d[1]) for d in dataset)) return tokens
[ "Returns a set of all tokens occurring the evaluation datasets." ]
Please provide a description of the function:def get_analogy_task_tokens(args): tokens = set() for _, _, dataset in iterate_analogy_datasets(args): tokens.update( itertools.chain.from_iterable( (d[0], d[1], d[2], d[3]) for d in dataset)) return tokens
[ "Returns a set of all tokens occuring the evaluation datasets." ]
Please provide a description of the function:def evaluate_similarity(args, token_embedding, ctx, logfile=None, global_step=0): results = [] for similarity_function in args.similarity_functions: evaluator = nlp.embedding.evaluation.WordEmbeddingSimilarity( idx_to...
[ "Evaluate on specified similarity datasets." ]
Please provide a description of the function:def evaluate_analogy(args, token_embedding, ctx, logfile=None, global_step=0): results = [] exclude_question_words = not args.analogy_dont_exclude_question_words for analogy_function in args.analogy_functions: evaluator = nlp.embedding.evaluation.Wor...
[ "Evaluate on specified analogy datasets.\n\n The analogy task is an open vocabulary task, make sure to pass a\n token_embedding with a sufficiently large number of supported tokens.\n\n " ]
Please provide a description of the function:def log_similarity_result(logfile, result): assert result['task'] == 'similarity' if not logfile: return with open(logfile, 'a') as f: f.write('\t'.join([ str(result['global_step']), result['task'], resul...
[ "Log a similarity evaluation result dictionary as TSV to logfile." ]
Please provide a description of the function:def get_model_loss(ctx, model, pretrained, dataset_name, dtype, ckpt_dir=None, start_step=None): # model model, vocabulary = nlp.model.get_model(model, dataset_name=dataset_name, ...
[ "Get model for pre-training." ]
Please provide a description of the function:def get_pretrain_dataset(data, batch_size, num_ctxes, shuffle, use_avg_len, num_buckets, num_parts=1, part_idx=0, prefetch=True): num_files = len(glob.glob(os.path.expanduser(data))) logging.debug('%d files found.', num_files) assert...
[ "create dataset for pretraining.", "create data loader based on the dataset chunk" ]
Please provide a description of the function:def get_dummy_dataloader(dataloader, target_shape): data_iter = enumerate(dataloader) _, data_batch = next(data_iter) logging.debug('Searching target batch shape: %s', target_shape) while data_batch[0].shape != target_shape: logging.debug('Skip b...
[ "Return a dummy data loader which returns a fixed data batch of target shape" ]
Please provide a description of the function:def save_params(step_num, model, trainer, ckpt_dir): param_path = os.path.join(ckpt_dir, '%07d.params'%step_num) trainer_path = os.path.join(ckpt_dir, '%07d.states'%step_num) logging.info('[step %d] Saving checkpoints to %s, %s.', step_num, ...
[ "Save the model parameter, marked by step_num." ]
Please provide a description of the function:def log(begin_time, running_num_tks, running_mlm_loss, running_nsp_loss, step_num, mlm_metric, nsp_metric, trainer, log_interval): end_time = time.time() duration = end_time - begin_time throughput = running_num_tks / duration / 1000.0 running_ml...
[ "Log training progress." ]
Please provide a description of the function:def split_and_load(arrs, ctx): assert isinstance(arrs, (list, tuple)) # split and load loaded_arrs = [mx.gluon.utils.split_and_load(arr, ctx, even_split=False) for arr in arrs] return zip(*loaded_arrs)
[ "split and load arrays to a list of contexts" ]
Please provide a description of the function:def forward(data, model, mlm_loss, nsp_loss, vocab_size, dtype): (input_id, masked_id, masked_position, masked_weight, \ next_sentence_label, segment_id, valid_length) = data num_masks = masked_weight.sum() + 1e-8 valid_length = valid_length.reshape(-1)...
[ "forward computation for evaluation" ]
Please provide a description of the function:def evaluate(data_eval, model, nsp_loss, mlm_loss, vocab_size, ctx, log_interval, dtype): mlm_metric = MaskedAccuracy() nsp_metric = MaskedAccuracy() mlm_metric.reset() nsp_metric.reset() eval_begin_time = time.time() begin_time = time.time() ...
[ "Evaluation function." ]
Please provide a description of the function:def get_argparser(): parser = argparse.ArgumentParser(description='BERT pretraining example.') parser.add_argument('--num_steps', type=int, default=20, help='Number of optimization steps') parser.add_argument('--num_buckets', type=int, default=1, ...
[ "Argument parser" ]
Please provide a description of the function:def _cache_dataset(dataset, prefix): if not os.path.exists(_constants.CACHE_PATH): os.makedirs(_constants.CACHE_PATH) src_data = np.concatenate([e[0] for e in dataset]) tgt_data = np.concatenate([e[1] for e in dataset]) src_cumlen = np.cumsum([0]...
[ "Cache the processed npy dataset the dataset into a npz\n\n Parameters\n ----------\n dataset : SimpleDataset\n file_path : str\n " ]
Please provide a description of the function:def load_translation_data(dataset, bleu, args): src_lang, tgt_lang = args.src_lang, args.tgt_lang if dataset == 'IWSLT2015': common_prefix = 'IWSLT2015_{}_{}_{}_{}'.format(src_lang, tgt_lang, args.sr...
[ "Load translation dataset\n\n Parameters\n ----------\n dataset : str\n args : argparse result\n\n Returns\n -------\n\n " ]
Please provide a description of the function:def make_dataloader(data_train, data_val, data_test, args, use_average_length=False, num_shards=0, num_workers=8): data_train_lengths = get_data_lengths(data_train) data_val_lengths = get_data_lengths(data_val) data_test_lengths = get_dat...
[ "Create data loaders for training/validation/test." ]
Please provide a description of the function:def run(self): random.seed(self.seed) np.random.seed(self.np_seed) if not isinstance(self, multiprocessing.Process): # Calling mxnet methods in a subprocess will raise an exception if # mxnet is built with GPU support ...
[ "Method representing the process’s activity." ]
Please provide a description of the function:def forward(self, inputs, states=None): # pylint: disable=arguments-differ batch_size = inputs.shape[self._batch_axis] skip_states = states is None if skip_states: states = self.cell.begin_state(batch_size, ctx=inputs.context) ...
[ "Defines the forward computation. Arguments can be either\n :py:class:`NDArray` or :py:class:`Symbol`." ]
Please provide a description of the function:def hybrid_forward(self, F, words, weight): #pylint: disable=arguments-differ embeddings = F.sparse.dot(words, weight) return embeddings
[ "Compute embedding of words in batch.\n\n Parameters\n ----------\n words : mx.nd.NDArray\n Array of token indices.\n\n " ]
Please provide a description of the function:def load_fasttext_format(cls, path, ctx=cpu(), **kwargs): with open(path, 'rb') as f: new_format, dim, bucket, minn, maxn, = cls._read_model_params(f) idx_to_token = cls._read_vocab(f, new_format) dim, matrix = cls._read_v...
[ "Create an instance of the class and load weights.\n\n Load the weights from the fastText binary format created by\n https://github.com/facebookresearch/fastText\n\n Parameters\n ----------\n path : str\n Path to the .bin model file.\n ctx : mx.Context, default m...