debugged

.gitignore

@@ -0,0 +1,164 @@
+.vscode
+
+# 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/
+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
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# 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
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .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
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/

docred.py

@@ -83,6 +83,7 @@ class docred(evaluate.Metric):
             "evidence": datasets.Sequence(datasets.Sequence(datasets.Value("int32"))),
         },
     }
+    eps = 1e-12
 
     def _info(self):
 
@@ -116,9 +117,9 @@ class docred(evaluate.Metric):
             labels = self._convert_labels_to_list(data["labels"])
             for label in labels:
                 rel = label["relation_id"]
-                for n1 in vertexSet[label["head"]]["name"]:
-                    for n2 in vertexSet[label["tail"]]["name"]:
-                        facts.add((n1, n2, rel))
+                for n1 in vertexSet[label["head"]]:
+                    for n2 in vertexSet[label["tail"]]:
+                        facts.add((n1["name"], n2["name"], rel))
         return facts
 
     def _convert_to_relation_set(self, data):
@@ -198,15 +199,15 @@ class docred(evaluate.Metric):
                 # if in_train_distant:
                 #     correct_in_train_distant += 1
 
-        re_p = 1.0 * correct_re / (len(submission_answer) + 1e-5)
-        re_r = 1.0 * correct_re / (tot_relations + 1e-5)
+        re_p = 1.0 * correct_re / (len(submission_answer) + self.eps)
+        re_r = 1.0 * correct_re / (tot_relations + self.eps)
         if re_p + re_r == 0:
             re_f1 = 0
         else:
             re_f1 = 2.0 * re_p * re_r / (re_p + re_r)
 
         evi_p = 1.0 * correct_evidence / pred_evi if pred_evi > 0 else 0
-        evi_r = 1.0 * correct_evidence / tot_evidences
+        evi_r = 1.0 * correct_evidence / (tot_evidences + self.eps)
         if evi_p + evi_r == 0:
             evi_f1 = 0
         else:
@@ -215,10 +216,10 @@ class docred(evaluate.Metric):
         re_p_ignore_train_annotated = (
             1.0
             * (correct_re - correct_in_train_annotated)
-            / (len(submission_answer) - correct_in_train_annotated + 1e-5)
+            / (len(submission_answer) - correct_in_train_annotated + self.eps)
         )
         # re_p_ignore_train = (
-        #     1.0 * (correct_re - correct_in_train_distant) / (len(submission_answer) - correct_in_train_distant + 1e-5)
+        #     1.0 * (correct_re - correct_in_train_distant) / (len(submission_answer) - correct_in_train_distant + self.eps)
         # )
 
         if re_p_ignore_train_annotated + re_r == 0:

sample.py

@@ -3,8 +3,17 @@ import evaluate
 
 from docred import docred
 
-train_data = datasets.load_dataset("docred", split="train_annotated[:10]")
-data = datasets.load_dataset("docred", split="validation[:10]")
+train_data = datasets.load_dataset("docred", split="train_annotated[:100]").to_list()
+pred_data = datasets.load_dataset("docred", split="validation[:10]").to_list()
+gold_data = datasets.load_dataset("docred", split="validation[:10]").to_list()
 metric = docred()
 
-print(metric.compute(predictions=data.to_list(), references=data.to_list()))
+# gold_data[0]["labels"] = {k: [] for k, v in pred_data[0]["labels"].items()}
+
+# for i in range(len(gold_data)):
+#     gold_data[i]["labels"] = {k: [] for k, v in pred_data[i]["labels"].items()}
+
+for i in range(len(pred_data)):
+    pred_data[i]["labels"] = {k: [] for k, v in pred_data[i]["labels"].items()}
+
+print(metric.compute(predictions=pred_data, references=gold_data, train_data=train_data))