poincare_train module¶
-
poincare_train.poincare_closer_then(poincare_model, word1, word2)[source]¶ Return the list of words closer to word1 in comparison with word2
Parameters: - word1 (str) – first word
- word2 (str) – second word
Returns: The list of segmented words
Return type: closer_list (list)
-
poincare_train.poincare_closest_child(poincare_model, word)[source]¶ Return the closet child node for a given word
Parameters: word (str) – arbitrary word Returns: The closest child word in Wordnet format Return type: child_word (str)
-
poincare_train.poincare_closest_parent(poincare_model, word)[source]¶ Return the closet parent node for a given word
Parameters: word (str) – arbitrary word Returns: The closest parent word in Wordnet format Return type: child_word (str)
-
poincare_train.poincare_simmilar(poincare_model, word)[source]¶ Return the list of words closest to word
Parameters: - poincare_model (model object) – The trained poincare model to use
- word (str) – The word used for finding similar words list
Returns: The list of similar words
Return type: most_simmilar_set (list)
-
poincare_train.poincare_train(hypertouple_dataset, size=2, burn_in=0, epochs=5, print_freq=100)[source]¶ Train a poincare embedding
Parameters: - hypertouple_dataset (list) – The hypertouple dataset to feed for training
- size (int) – size of model
- burn_in (int) – Burnin identifier
- epochs (int) – Number of epochs to train
- print_freq (int) – Update frequency number
Returns: The trained Poincare Model
Return type: poincare_model (model object)