Functions — github.com/allenai/commonsense-kg-completion

hub / github.com/allenai/commonsense-kg-completion / functions

Functions174 in github.com/allenai/commonsense-kg-completion

⨍ Functions 174◇ Types & classes 33

Method__init__

(self, node1, node2, relation, label, weight, uri)

src/graph.py:177

Method__init__

(self, num_nodes, num_rels, args, use_cuda=False)

src/model.py:14

Method__init__

(self, num_nodes, h_dim, dataset=None, init_bert=None)

src/model.py:162

Method__init__

(self, corpus_path, tokenizer, seq_len, encoding="utf-8", corpus_lines=None, on_memory=True)

lm_finetuning/simple_lm_finetuning.py:42

Method__init__

Constructs a InputExample. Args: guid: Unique id for the example. tokens_a: string. The untokenized text of the first

lm_finetuning/simple_lm_finetuning.py:236

Method__init__

(self, input_ids, input_mask, segment_ids, is_next, lm_label_ids)

lm_finetuning/simple_lm_finetuning.py:257

Method__len__

(self)

lm_finetuning/simple_lm_finetuning.py:107

lm_finetuning/simple_lm_finetuning.py:632

Methodbuild_hidden_layer

(self, idx)

src/model.py:261

Methodbuild_input_layer

(self, dataset, input_layer="lookup")

src/model.py:242

Functionbuild_test_graph

(num_nodes, num_rels, edges)

src/utils.py:177

Methodcompute_attn_weights

()

src/layers.py:200

Methodcompute_attn_weights

()

src/layers.py:318

Functionconvert_edges_to_examples

(edges, labels, network)

src/bert_feature_extractor.py:128

Functioncreate_entity_dicts

(all_tuples, num_rels, sim_relations=False)

src/reader_utils.py:92

Methodcreate_features

src/reader_utils.py:28

Functioncreate_word_vocab

(network)

src/reader_utils.py:8

(self, g, epoch=None)

src/model.py:127

Methodfind_all_connections

:param relation: :return: list of all edges representing this relation

src/graph.py:111

Methodforward

(self, embedding, e1, rel)

src/decoder.py:25

Methodforward

(self, embedding, e1, rel)

src/decoder.py:59

Methodforward

(self, embedding, triplets)

src/decoder.py:119

Methodforward

(self, e1, rel, target)

Methodgen_negative_examples

(self, tgt_size=None, sampling_type="random")

src/reader.py:36

Functiongenerate_sampled_graph_and_labels

Get training graph and signals First perform edge neighborhood sampling on graph, then perform negative sampling to generate negative samples

src/utils.py:90

Functionget_adj_and_degrees

Get adjacency list and degrees of the graph

src/utils.py:12

Methodget_neighbors

:param node: :return:

:param node1: :param node2: :return:

Compute messages only from source node features

src/layers.py:154

Methodmsg_func

Compute messages only from source node features

src/layers.py:189

Methodmsg_func

Compute messages only from source node features

src/layers.py:230

Methodmsg_func

Compute messages only from source node features

src/layers.py:265

Methodmsg_func

Compute messages only from source node features

src/layers.py:308

Methodnode_exists

:param node: node to check :return: Boolean value

src/graph.py:101

Functionperturb_and_get_rank

Perturb one element in the triplets

src/evaluation_utils.py:41

Functionperturb_and_get_rank_conv

Perturb one element in the triplets for a convolution-based decoder

src/evaluation_utils.py:75

Functionplot_degree_mrr

(node_ranks)

src/evaluation_utils.py:257

Functionprepare_batch_dgl

(vocab, test_network, train_network)

src/reader_utils.py:80

Functionprepare_batch_nodes

(vocab, batch)

src/reader_utils.py:63

Functionpreprocess_atomic_sentence

(sent)

src/reader_utils.py:126

Functionranking_and_hits

(test_graph, model, test_triplets, e1_to_multi_e2, network, fusion="graph-only", sim_rela

src/evaluation_utils.py:105

Methodread_network

(self, data_path, split="trn")

← previous101–174 of 174, ranked by callers