Zhibin Lu
commited on
Commit
•
50c753d
1
Parent(s):
777dbcc
some comments
Browse files- configuration_vgcn_bert.py +0 -1
- modeling_vgcn_bert.py +3 -1
configuration_vgcn_bert.py
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
# coding=utf-8
|
| 2 |
-
# Copyright 2019-present, the HuggingFace Inc. team, The Google AI Language Team and Facebook, Inc.
|
| 3 |
#
|
| 4 |
# Licensed under the Apache License, Version 2.0 (the "License");
|
| 5 |
# you may not use this file except in compliance with the License.
|
|
|
|
| 1 |
# coding=utf-8
|
|
|
|
| 2 |
#
|
| 3 |
# Licensed under the Apache License, Version 2.0 (the "License");
|
| 4 |
# you may not use this file except in compliance with the License.
|
modeling_vgcn_bert.py
CHANGED
|
@@ -404,13 +404,15 @@ def _build_predefined_graph(
|
|
| 404 |
|
| 405 |
|
| 406 |
# TODO: build knowledge graph from a list of RDF triples
|
| 407 |
-
# def _build_knowledge_graph
|
| 408 |
|
| 409 |
|
| 410 |
class WordGraphBuilder:
|
| 411 |
"""
|
| 412 |
Word graph based on adjacency matrix, construct from text samples or pre-defined word-pair relations
|
| 413 |
|
|
|
|
|
|
|
|
|
|
| 414 |
Params:
|
| 415 |
`rows`: List[str] of text samples, or pre-defined word-pair relations: List[Tuple[str, str, float]]
|
| 416 |
`tokenizer`: The same pretrained tokenizer that is used for the model late.
|
|
|
|
| 404 |
|
| 405 |
|
| 406 |
# TODO: build knowledge graph from a list of RDF triples
|
|
|
|
| 407 |
|
| 408 |
|
| 409 |
class WordGraphBuilder:
|
| 410 |
"""
|
| 411 |
Word graph based on adjacency matrix, construct from text samples or pre-defined word-pair relations
|
| 412 |
|
| 413 |
+
You may (or not) first preprocess the text before build the graph,
|
| 414 |
+
e.g. Stopword removal, String cleaning, Stemming, Nomolization, Lemmatization
|
| 415 |
+
|
| 416 |
Params:
|
| 417 |
`rows`: List[str] of text samples, or pre-defined word-pair relations: List[Tuple[str, str, float]]
|
| 418 |
`tokenizer`: The same pretrained tokenizer that is used for the model late.
|