Upload 2 files

gitignore

@@ -0,0 +1,143 @@
+# dev files
+*.cache
+*.dev.py
+state_dict/
+TAD*/
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+*.pyc
+tests/
+*.result.json
+.idea/
+
+# Embedding
+glove.840B.300d.txt
+glove.42B.300d.txt
+glove.twitter.27B.txt
+
+# project main files
+release_note.json
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+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 training_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/
+
+# 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
+
+# celery beat schedule file
+celerybeat-schedule
+
+# 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/
+.DS_Store
+examples/.DS_Store

utils.py

@@ -0,0 +1,234 @@
+import random
+from difflib import Differ
+
+from textattack.attack_recipes import BAEGarg2019
+from textattack.datasets import Dataset
+from textattack.models.wrappers import HuggingFaceModelWrapper
+from findfile import find_files
+from flask import Flask
+from textattack import Attacker
+
+
+class ModelWrapper(HuggingFaceModelWrapper):
+    def __init__(self, model):
+        self.model = model  # pipeline = pipeline
+
+    def __call__(self, text_inputs, **kwargs):
+        outputs = []
+        for text_input in text_inputs:
+            raw_outputs = self.model.infer(text_input, print_result=False, **kwargs)
+            outputs.append(raw_outputs["probs"])
+        return outputs
+
+
+class SentAttacker:
+    def __init__(self, model, recipe_class=BAEGarg2019):
+        model = model
+        model_wrapper = ModelWrapper(model)
+
+        recipe = recipe_class.build(model_wrapper)
+        # WordNet defaults to english. Set the default language to French ('fra')
+
+        # recipe.transformation.language = "en"
+
+        _dataset = [("", 0)]
+        _dataset = Dataset(_dataset)
+
+        self.attacker = Attacker(recipe, _dataset)
+
+
+def diff_texts(text1, text2):
+    d = Differ()
+    return [
+        (token[2:], token[0] if token[0] != " " else None)
+        for token in d.compare(text1, text2)
+    ]
+
+
+def get_ensembled_tad_results(results):
+    target_dict = {}
+    for r in results:
+        target_dict[r["label"]] = (
+            target_dict.get(r["label"]) + 1 if r["label"] in target_dict else 1
+        )
+
+    return dict(zip(target_dict.values(), target_dict.keys()))[
+        max(target_dict.values())
+    ]
+
+
+
+def get_sst2_example():
+    filter_key_words = [
+        ".py",
+        ".md",
+        "readme",
+        "log",
+        "result",
+        "zip",
+        ".state_dict",
+        ".model",
+        ".png",
+        "acc_",
+        "f1_",
+        ".origin",
+        ".adv",
+        ".csv",
+    ]
+
+    dataset_file = {"train": [], "test": [], "valid": []}
+    dataset = "sst2"
+    search_path = "./"
+    task = "text_defense"
+    dataset_file["test"] += find_files(
+        search_path,
+        [dataset, "test", task],
+        exclude_key=[".adv", ".org", ".defense", ".inference", "train."]
+                    + filter_key_words,
+    )
+
+    for dat_type in ["test"]:
+        data = []
+        label_set = set()
+        for data_file in dataset_file[dat_type]:
+            with open(data_file, mode="r", encoding="utf8") as fin:
+                lines = fin.readlines()
+                for line in lines:
+                    text, label = line.split("$LABEL$")
+                    text = text.strip()
+                    label = int(label.strip())
+                    data.append((text, label))
+                    label_set.add(label)
+        return random.choice(data)
+
+
+def get_agnews_example():
+    filter_key_words = [
+        ".py",
+        ".md",
+        "readme",
+        "log",
+        "result",
+        "zip",
+        ".state_dict",
+        ".model",
+        ".png",
+        "acc_",
+        "f1_",
+        ".origin",
+        ".adv",
+        ".csv",
+    ]
+
+    dataset_file = {"train": [], "test": [], "valid": []}
+    dataset = "agnews"
+    search_path = "./"
+    task = "text_defense"
+    dataset_file["test"] += find_files(
+        search_path,
+        [dataset, "test", task],
+        exclude_key=[".adv", ".org", ".defense", ".inference", "train."]
+                    + filter_key_words,
+    )
+    for dat_type in ["test"]:
+        data = []
+        label_set = set()
+        for data_file in dataset_file[dat_type]:
+            with open(data_file, mode="r", encoding="utf8") as fin:
+                lines = fin.readlines()
+                for line in lines:
+                    text, label = line.split("$LABEL$")
+                    text = text.strip()
+                    label = int(label.strip())
+                    data.append((text, label))
+                    label_set.add(label)
+        return random.choice(data)
+
+
+def get_amazon_example():
+    filter_key_words = [
+        ".py",
+        ".md",
+        "readme",
+        "log",
+        "result",
+        "zip",
+        ".state_dict",
+        ".model",
+        ".png",
+        "acc_",
+        "f1_",
+        ".origin",
+        ".adv",
+        ".csv",
+    ]
+
+    dataset_file = {"train": [], "test": [], "valid": []}
+    dataset = "amazon"
+    search_path = "./"
+    task = "text_defense"
+    dataset_file["test"] += find_files(
+        search_path,
+        [dataset, "test", task],
+        exclude_key=[".adv", ".org", ".defense", ".inference", "train."]
+                    + filter_key_words,
+    )
+
+    for dat_type in ["test"]:
+        data = []
+        label_set = set()
+        for data_file in dataset_file[dat_type]:
+            with open(data_file, mode="r", encoding="utf8") as fin:
+                lines = fin.readlines()
+                for line in lines:
+                    text, label = line.split("$LABEL$")
+                    text = text.strip()
+                    label = int(label.strip())
+                    data.append((text, label))
+                    label_set.add(label)
+        return random.choice(data)
+
+
+def get_imdb_example():
+    filter_key_words = [
+        ".py",
+        ".md",
+        "readme",
+        "log",
+        "result",
+        "zip",
+        ".state_dict",
+        ".model",
+        ".png",
+        "acc_",
+        "f1_",
+        ".origin",
+        ".adv",
+        ".csv",
+    ]
+
+    dataset_file = {"train": [], "test": [], "valid": []}
+    dataset = "imdb"
+    search_path = "./"
+    task = "text_defense"
+    dataset_file["test"] += find_files(
+        search_path,
+        [dataset, "test", task],
+        exclude_key=[".adv", ".org", ".defense", ".inference", "train."]
+                    + filter_key_words,
+    )
+
+    for dat_type in ["test"]:
+        data = []
+        label_set = set()
+        for data_file in dataset_file[dat_type]:
+            with open(data_file, mode="r", encoding="utf8") as fin:
+                lines = fin.readlines()
+                for line in lines:
+                    text, label = line.split("$LABEL$")
+                    text = text.strip()
+                    label = int(label.strip())
+                    data.append((text, label))
+                    label_set.add(label)
+        return random.choice(data)
+