Depend on main REL and get tests to pass [do not merge]

.gitignore

@@ -0,0 +1,134 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+coverage.json
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+pyvenv.cfg
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# vscode
+.vscode/
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/

tool/conversational_entity_linking.ipynb → crel/conversational_entity_linking.ipynb

@@ -112,7 +112,7 @@
    "hash": "a53556614ea0d196de1dd499c6cd4b1019f00d4a13a34e20ba99029df2a473df"
   },
   "kernelspec": {
-   "display_name": "Python 3.8.13 ('220429_conel22_github')",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -126,9 +126,8 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.13"
-  },
-  "orig_nbformat": 4
+   "version": "3.9.15"
+  }
  },
  "nbformat": 4,
  "nbformat_minor": 2

crel/setup.cfg

@@ -0,0 +1,22 @@
+[metadata]
+name = crel
+
+[options]
+zip_safe = False
+packages = find_namespace:
+package_dir =
+    = src
+include_package_data = True
+install_requires =
+    radboud-el
+    spacy
+
+[options.extras_require]
+develop =
+	pytest
+
+[options.packages.find]
+where = src
+
+[tool:pytest]
+testpaths = tests

crel/setup.py

@@ -0,0 +1,2 @@
+from setuptools import setup
+setup()

tool/conv_el.py → crel/src/crel/conv_el.py

@@ -1,19 +1,22 @@
 import sys
-sys.path.append('s2e_pe')
-import pe_data
+from .s2e_pe import pe_data
 import importlib
-from bert_md import BERT_MD
-from rel_ed import REL_ED
-from pe import EEMD, PEMD
+from .bert_md import BERT_MD
+from .rel_ed import REL_ED
+from .s2e_pe.pe import EEMD, PEMD
 
 class ConvEL():
-    def __init__(self, threshold=0):
+    def __init__(self, threshold=0, config=None):
         self.threshold = threshold
 
-        conf = self.ConfigConvEL()
+        if not config:
+            config = {}
+
+        conf = self.ConfigConvEL(**config)
+
         self.bert_md = BERT_MD(conf.file_pretrained)
         self.rel_ed = REL_ED(conf.base_url, conf.wiki_version)
-        self.eemd = EEMD()
+        self.eemd = EEMD(config=config)
         self.pemd = PEMD()
 
         self.preprocess = pe_data.PreProcess()
@@ -24,13 +27,18 @@ def __init__(self, threshold=0):
         self.ment2ent = {} # This will be used for PE Linking
 
     class ConfigConvEL():
-        def __init__(self):
+        def __init__(self, 
+                file_pretrained = './bert_conv-td', 
+                base_url = './rel_conv_project_folder',
+                wiki_version='wiki_2019',
+                **kwargs
+            ):
             # MD
-            self.file_pretrained = './bert_conv-td'
+            self.file_pretrained = file_pretrained
 
             # ED
-            self.base_url = './rel_conv_project_folder'
-            self.wiki_version = "wiki_2019"
+            self.base_url = base_url
+            self.wiki_version = wiki_version
 
             # NOTE: PE Config is in EEMD class
 

tool/rel_ed.py → crel/src/crel/rel_ed.py

@@ -1,5 +1,4 @@
 import sys
-sys.path.append('./REL')
 from REL.entity_disambiguation import EntityDisambiguation
 from REL.utils import process_results
 from REL.mention_detection import MentionDetection

tool/s2e_pe/data.py → crel/src/crel/s2e_pe/data.py

@@ -6,8 +6,8 @@
 
 import torch
 
-from consts import SPEAKER_START, SPEAKER_END, NULL_ID_FOR_COREF
-from utils import flatten_list_of_lists
+from .consts import SPEAKER_START, SPEAKER_END, NULL_ID_FOR_COREF
+from .utils import flatten_list_of_lists
 from torch.utils.data import Dataset
 
 CorefExample = namedtuple("CorefExample", ["token_ids", "clusters"])

tool/s2e_pe/modeling.py → crel/src/crel/s2e_pe/modeling.py

@@ -5,7 +5,7 @@
     from transformers.modeling_bert import ACT2FN
 except: # If you use `jupyterlab-debugger`
     from transformers.models.bert.modeling_bert import ACT2FN
-from utils import extract_clusters, extract_mentions_to_predicted_clusters_from_clusters, mask_tensor #, ce_get_start_end_subtoken_num
+from .utils import extract_clusters, extract_mentions_to_predicted_clusters_from_clusters, mask_tensor #, ce_get_start_end_subtoken_num
 import os
 import json
 

tool/s2e_pe/pe.py → crel/src/crel/s2e_pe/pe.py

@@ -10,14 +10,14 @@
     spacy.cli.download("en_core_web_md")
     import en_core_web_md
     nlp = en_core_web_md.load()
-from pe_data import PreProcess # to use get_span()
+from .pe_data import PreProcess # to use get_span()
 
 # EEMD
-import data
+from . import data
 import torch
 from transformers import AutoConfig, AutoTokenizer, LongformerConfig
-from modeling import S2E
-from coref_bucket_batch_sampler import BucketBatchSampler
+from .modeling import S2E
+from .coref_bucket_batch_sampler import BucketBatchSampler
 
 
 class PEMD():
@@ -119,16 +119,22 @@ class EEMD():
     """Find corresponding explicit entity mention using s2e-coref-based method
     """
 
-    def __init__(self):
-        self.conf = self.Config()
+    def __init__(self, config=None):
+        if not config:
+            config = {}
+
+        self.conf = self.Config(**config)
         self.model = self._read_model()
 
     class Config():
         """Inner class for config
         """
-        def __init__(self):
+        def __init__(self, 
+                s2e_pe_model='./s2e_pe/model/s2e_ast_onto',
+                **kwargs
+            ):
             self.max_seq_length = 4096
-            self.model_name_or_path = './s2e_pe/model/s2e_ast_onto'
+            self.model_name_or_path = s2e_pe_model
             self.max_total_seq_len = 4096
             self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
             # self.device = torch.device("cpu") # TMP

tool/s2e_pe/utils.py → crel/src/crel/s2e_pe/utils.py

@@ -6,7 +6,7 @@
 import torch
 import numpy as np
 
-from consts import NULL_ID_FOR_COREF
+from .consts import NULL_ID_FOR_COREF
 
 
 def flatten_list_of_lists(lst):