RuleModel¶
IterE¶
- class neuralkg.model.RuleModel.IterE.IterE(args, train_sampler, test_sampler)[source]¶
Bases:
neuralkg.model.RuleModel.model.ModelIteratively Learning Embeddings and Rules for Knowledge Graph Reasoning. (WWW’19) (IterE).
- args¶
Model configuration parameters.
- epsilon¶
Caculate embedding_range.
- margin¶
Caculate embedding_range and loss.
- embedding_range¶
Uniform distribution range.
- ent_emb¶
Entity embedding, shape:[num_ent, emb_dim].
- rel_emb¶
Relation_embedding, shape:[num_rel, emb_dim].
- update_train_triples(epoch=0, update_per=10)[source]¶
add the new triples from axioms to training triple
- split_embedding(embedding)[source]¶
split embedding
- Parameters
embedding – embeddings need to be splited, shape:[None, dim].
- Returns
The similrity between two matrices.
- Return type
probability
- sim(head=None, tail=None, arity=None)[source]¶
calculate the similrity between two matrices
- Parameters
head – embeddings of head, shape:[batch_size, dim].
tail – embeddings of tail, shape:[batch_size, dim] or [1, dim].
arity – 1，2 or 3
- Returns
The similrity between two matrices.
- Return type
probability
- run_axiom_probability()[source]¶
this function is used to generate a probality for each axiom in axiom pool
- update_valid_axioms(input)[source]¶
this function is used to select high probability axioms as valid axioms and record their scores
- generate_new_train_triples()[source]¶
The function is to updata new train triples and used after each training epoch end
- Returns
The new training dataset (triples).
- Return type
self.train_sampler.train_triples
- get_rule(rel2id)[source]¶
Get rule for rule_base KGE models, such as ComplEx_NNE model. Get rule and confidence from _cons.txt file. Update:
(rule_p, rule_q): Rule. confidence: The confidence of rule.
- init_emb()[source]¶
Initialize the entity and relation embeddings in the form of a uniform distribution.
- score_func(head_emb, relation_emb, tail_emb, mode)[source]¶
Calculating the score of triples.
The formula for calculating the score is DistMult.
- Parameters
head_emb – The head entity embedding.
relation_emb – The relation embedding.
tail_emb – The tail entity embedding.
mode – Choose head-predict or tail-predict.
- Returns
The score of triples.
- Return type
score
- forward(triples, negs=None, mode='single')[source]¶
The functions used in the training phase
- Parameters
triples – The triples ids, as (h, r, t), shape:[batch_size, 3].
negs – Negative samples, defaults to None.
mode – Choose head-predict or tail-predict, Defaults to ‘single’.
- Returns
The score of triples.
- Return type
score
RugE¶
- class neuralkg.model.RuleModel.RugE.RugE(args)[source]¶
Bases:
neuralkg.model.RuleModel.model.ModelKnowledge Graph Embedding with Iterative Guidance from Soft Rules (RugE), which is a novel paradigm of KG embedding with iterative guidance from soft rules.
- args¶
Model configuration parameters.
- epsilon¶
Caculate embedding_range.
- margin¶
Caculate embedding_range and loss.
- embedding_range¶
Uniform distribution range.
- ent_emb¶
Entity embedding, shape:[num_ent, emb_dim].
- rel_emb¶
Relation_embedding, shape:[num_rel, emb_dim].
- init_emb()[source]¶
Initialize the entity and relation embeddings in the form of a uniform distribution.
- score_func(head_emb, relation_emb, tail_emb, mode)[source]¶
Calculating the score of triples.
The formula for calculating the score is \(Re(< wr, es, e¯o >)\)
- Parameters
head_emb – The head entity embedding.
relation_emb – The relation embedding.
tail_emb – The tail entity embedding.
mode – Choose head-predict or tail-predict.
- Returns
The score of triples.
- Return type
score
- forward(triples, negs=None, mode='single')[source]¶
The functions used in the training phase
- Parameters
triples – The triples ids, as (h, r, t), shape:[batch_size, 3].
negs – Negative samples, defaults to None.
mode – Choose head-predict or tail-predict, Defaults to ‘single’.
- Returns
The score of triples.
- Return type
score
ComplEx_NNE_AER¶
- class neuralkg.model.RuleModel.ComplEx_NNE_AER.ComplEx_NNE_AER(args, rel2id)[source]¶
Bases:
neuralkg.model.RuleModel.model.ModelImproving Knowledge Graph Embedding Using Simple Constraints (/ComplEx-NNE_AER), which examines non-negativity constraints on entity representations and approximate entailment constraints on relation representations.
- args¶
Model configuration parameters.
- epsilon¶
Caculate embedding_range.
- margin¶
Caculate embedding_range and loss.
- embedding_range¶
Uniform distribution range.
- ent_emb¶
Entity embedding, shape:[num_ent, emb_dim].
- rel_emb¶
Relation_embedding, shape:[num_rel, emb_dim].
- get_rule(rel2id)[source]¶
Get rule for rule_base KGE models, such as ComplEx_NNE model. Get rule and confidence from _cons.txt file. Update:
(rule_p, rule_q): Rule. confidence: The confidence of rule.
- init_emb()[source]¶
Initialize the entity and relation embeddings in the form of a uniform distribution.
- score_func(head_emb, relation_emb, tail_emb, mode)[source]¶
Calculating the score of triples.
The formula for calculating the score is \(Re(< wr, es, e¯o >)\)
- Parameters
head_emb – The head entity embedding.
relation_emb – The relation embedding.
tail_emb – The tail entity embedding.
mode – Choose head-predict or tail-predict.
- Returns
The score of triples.
- Return type
score
- forward(triples, negs=None, mode='single')[source]¶
The functions used in the training phase
- Parameters
triples – The triples ids, as (h, r, t), shape:[batch_size, 3].
negs – Negative samples, defaults to None.
mode – Choose head-predict or tail-predict, Defaults to ‘single’.
- Returns
The score of triples.
- Return type
score