Merge pull request #575 from srujanjoshi/documentation_fix

Fixed issues with Documentation + Added Copy Button for Code Blocks
QData · Nov 13, 2021 · bfc3468 · bfc3468
2 parents b05c555 + 59bd582
commit bfc3468
Show file tree

Hide file tree

Showing 26 changed files with 75 additions and 40 deletions.
diff --git a/docs/0_get_started/basic-Intro.rst b/docs/0_get_started/basic-Intro.rst
@@ -19,9 +19,9 @@ This is a great question, and one we get a lot. First of all, almost everything
 
 TextAttack does three things very well:
 
-1. Adversarial attacks (Python: ``textattack.shared.Attack``, Bash: ``textattack attack``)
+1. Adversarial attacks (Python: ``textattack.Attack``, Bash: ``textattack attack``)
 2. Data augmentation (Python: ``textattack.augmentation.Augmenter``, Bash: ``textattack augment``)
-3. Model training (Python: ``textattack.commands.train.*``, Bash: ``textattack train``)
+3. Model training (Python: ``textattack.Trainer``, Bash: ``textattack train``)
 
 Adversarial training can be achieved as a combination of [1] and/or [2] with [3] (via ``textattack train --attack``). To see all this in action, see :ref:`the TextAttack End-to-End tutorial </2notebook/0_End_to_End.ipynb>`.
 

diff --git a/docs/0_get_started/quick_api_tour.rst b/docs/0_get_started/quick_api_tour.rst
@@ -29,7 +29,6 @@ Next, let's build the attack that we want to use. TextAttack provides prebuilt a
     >>> dataset = textattack.datasets.HuggingFaceDataset("imdb", split="test")
     >>> attack = textattack.attack_recipes.TextFoolerJin2019.build(model_wrapper)
     >>> # Attack 20 samples with CSV logging and checkpoint saved every 5 interval
-    >>> # Attack 20 samples with CSV logging and checkpoint saved every 5 interval
     >>> attack_args = textattack.AttackArgs(num_examples=20, log_to_csv="log.csv", checkpoint_interval=5, checkpoint_dir="checkpoints", disable_stdout=True)
     >>> attacker = textattack.Attacker(attack, dataset, attack_args)
     >>> attacker.attack_dataset()

diff --git a/docs/3recipes/models.md b/docs/3recipes/models.md
@@ -447,10 +447,10 @@ roberta-base-RTE                         |  natural language inference
 roberta-base-WNLI                        |  natural language inference                     |  sentence pairs                               |  binary                                             | <sub><sup> https://paperswithcode.com/sota/natural-language-inference-on-wnli </sub></sup>           |  https://huggingface.co/textattack/roberta-base-WNLI </sub></sup>
 xlnet-base-cased-RTE                     |  natural language inference                     |  sentence pairs (1 premise and 1 hypothesis)  |  binary(0=entailed/1=not entailed)                  | <sub><sup>  https://paperswithcode.com/sota/ </sub></sup>natural-language-inference-on-rte             | <sub><sup> https://huggingface.co/textattack/xlnet-base-cased-RTE </sub></sup>
 xlnet-base-cased-WNLI                    |  natural language inference                     |  sentence pairs                               |  binary                                             |  none yet                                                                      | <sub><sup> https://huggingface.co/textattack/xlnet-base-cased-WNLI </sub></sup>
-albert-base-v2-QQP                       |  paraphase similarity                           |  question pairs                               |  binary (1=similar/0=not similar)                   | <sub><sup> https://paperswithcode.com/sota/question-answering-on-quora-question-pairs  </sub></sup>  | <sub><sup> https://huggingface.co/textattack/albert-base-v2-QQP</sub></sup>
-bert-base-uncased-QQP                    |  paraphase similarity                           |  question pairs                               |  binary (1=similar/0=not similar)                   | <sub><sup> https://paperswithcode.com/sota/question-answering-on-quora-question-pairs  </sub></sup>  | <sub><sup> https://huggingface.co/textattack/bert-base-uncased-QQP </sub></sup>
+albert-base-v2-QQP                       |  paraphrase similarity                           |  question pairs                               |  binary (1=similar/0=not similar)                   | <sub><sup> https://paperswithcode.com/sota/question-answering-on-quora-question-pairs  </sub></sup>  | <sub><sup> https://huggingface.co/textattack/albert-base-v2-QQP</sub></sup>
+bert-base-uncased-QQP                    |  paraphrase similarity                           |  question pairs                               |  binary (1=similar/0=not similar)                   | <sub><sup> https://paperswithcode.com/sota/question-answering-on-quora-question-pairs  </sub></sup>  | <sub><sup> https://huggingface.co/textattack/bert-base-uncased-QQP </sub></sup>
 distilbert-base-uncased-QNLI             |  question answering/natural language inference  |  question/answer pairs                        |  binary (1=unanswerable/ 0=answerable)               | <sub><sup> https://paperswithcode.com/sota/natural-language-inference-on-qnli   </sub></sup>         | <sub><sup> https://huggingface.co/textattack/distilbert-base-uncased-QNLI </sub></sup>
-distilbert-base-cased-QQP                |  question answering/paraphase similarity        |  question pairs                               |  binary (1=similar/ 0=not similar)                   | <sub><sup> https://paperswithcode.com/sota/question-answering-on-quora-question-pairs  </sub></sup>  | <sub><sup> https://huggingface.co/textattack/distilbert-base-cased-QQP </sub></sup>
+distilbert-base-cased-QQP                |  question answering/paraphrase similarity        |  question pairs                               |  binary (1=similar/ 0=not similar)                   | <sub><sup> https://paperswithcode.com/sota/question-answering-on-quora-question-pairs  </sub></sup>  | <sub><sup> https://huggingface.co/textattack/distilbert-base-cased-QQP </sub></sup>
 albert-base-v2-STS-B                     |  semantic textual similarity                    |  sentence pairs                               |  similarity (0.0 to 5.0)                            | <sub><sup> https://paperswithcode.com/sota/semantic-textual-similarity-on-sts-benchmark </sub></sup> | <sub><sup> https://huggingface.co/textattack/albert-base-v2-STS-B </sub></sup>
 bert-base-uncased-MRPC                   |  semantic textual similarity                    |  sentence pairs                               |  binary (1=similar/0=not similar)                   |  none yet                                                                      | <sub><sup> https://huggingface.co/textattack/bert-base-uncased-MRPC </sub></sup>
 bert-base-uncased-STS-B                  |  semantic textual similarity                    |  sentence pairs                               |  similarity (0.0 to 5.0)                            |  none yet                                                                      | <sub><sup> https://huggingface.co/textattack/bert-base-uncased-STS-B </sub></sup>

diff --git a/docs/api/attack.rst b/docs/api/attack.rst
@@ -5,10 +5,10 @@ Attack
 ------------
 Attack is composed of four components:
 
-- `Goal Functions <../attacks/goal_function.html>`__ stipulate the goal of the attack, like to change the prediction score of a classification model, or to change all of the words in a translation output.
-- `Constraints <../attacks/constraint.html>`__ determine if a potential perturbation is valid with respect to the original input.
-- `Transformations <../attacks/transformation.html>`__ take a text input and transform it by inserting and deleting characters, words, and/or phrases.
-- `Search Methods <../attacks/search_method.html>`__ explore the space of possible **transformations** within the defined **constraints** and attempt to find a successful perturbation which satisfies the **goal function**.
+- :ref:`Goal Functions  <goal_functions>`: stipulate the goal of the attack, like to change the prediction score of a classification model, or to change all of the words in a translation output.
+- :ref:`Constraints <constraint>`: determine if a potential perturbation is valid with respect to the original input.
+- :ref:`Transformations  <transformations>`: take a text input and transform it by inserting and deleting characters, words, and/or phrases.
+- :ref:`Search Methods  <search_methods>`: explore the space of possible **transformations** within the defined **constraints** and attempt to find a successful perturbation which satisfies the **goal function**.
 
 The :class:`~textattack.Attack` class represents an adversarial attack composed of a goal function, search method, transformation, and constraints.
 

diff --git a/docs/api/goal_functions.rst b/docs/api/goal_functions.rst
@@ -11,36 +11,36 @@ GoalFunction
 
 ClassificationGoalFunction
 --------------------------
-.. autoclass:: textattack.goal_functions.ClassificationGoalFunction
+.. autoclass:: textattack.goal_functions.classification.ClassificationGoalFunction
    :members:
 
 TargetedClassification
 ----------------------
-.. autoclass:: textattack.goal_functions.TargetedClassification
+.. autoclass:: textattack.goal_functions.classification.TargetedClassification
    :members:
 
 UntargetedClassification
 ------------------------
-.. autoclass:: textattack.goal_functions.UntargetedClassification
+.. autoclass:: textattack.goal_functions.classification.UntargetedClassification
    :members:
 
 InputReduction
 --------------
-.. autoclass:: textattack.goal_functions.InputReduction
+.. autoclass:: textattack.goal_functions.classification.InputReduction
    :members:
 
 TextToTextGoalFunction
 -----------------------
-.. autoclass:: textattack.goal_functions.TextToTextGoalFunction
+.. autoclass:: textattack.goal_functions.text.TextToTextGoalFunction
    :members:
 
 MinimizeBleu
 -------------
-.. autoclass:: textattack.goal_functions.MinimizeBleu
+.. autoclass:: textattack.goal_functions.text.MinimizeBleu
    :members:
 
 NonOverlappingOutput
 ----------------------
-.. autoclass:: textattack.goal_functions.NonOverlappingOutput
+.. autoclass:: textattack.goal_functions.text.NonOverlappingOutput
    :members:
 
diff --git a/docs/api/transformations.rst b/docs/api/transformations.rst
@@ -0,0 +1,13 @@
+Transformations API Reference
+===============================
+
+Transformation
+----------------
+.. autoclass:: textattack.transformations.Transformation
+   :members:
+
+Composite Transformation
+-----------------------------
+.. autoclass:: textattack.transformations.CompositeTransformation
+    :members:
+
diff --git a/docs/conf.py b/docs/conf.py
@@ -17,7 +17,7 @@
 
 # -- Project information -----------------------------------------------------
 project = "TextAttack"
-copyright = "2020, UVA QData Lab"
+copyright = "2021, UVA QData Lab"
 author = "UVA QData Lab"
 
 # The full version, including alpha/beta/rc tags
@@ -39,6 +39,7 @@
     "sphinx.ext.autosummary",
     "sphinx.ext.napoleon",
     "sphinx_rtd_theme",
+    "sphinx_copybutton",
     # Enable .ipynb doc files
     "nbsphinx",
     # Enable .md doc files
@@ -94,3 +95,11 @@
 
 # Sort members by group
 autodoc_member_order = "groupwise"
+
+# -- Options for Sphinx Copy Button-------------------------------------------------
+
+# Exclude the prompt symbol ">>>" when copying text
+copybutton_prompt_text = ">>> "
+
+# Specify Line Continuation Character so that all the entire Line is copied
+copybutton_line_continuation_character = "\\"
diff --git a/docs/index.rst b/docs/index.rst
@@ -63,6 +63,7 @@ TextAttack Documentation
    Datasets <api/datasets.rst>
    GoalFunction <api/goal_functions.rst>
    Constraints <api/constraints.rst>
+   Transformations <api/transformations.rst>
    SearchMethod <api/search_methods.rst>
 
 

diff --git a/docs/requirements.txt b/docs/requirements.txt
@@ -1,6 +1,7 @@
 recommonmark==0.7.1
 Sphinx==4.1.2
 sphinx-autobuild==2021.3.14
+sphinx-copybutton==0.4.0
 sphinx-markdown-tables==0.0.15
 sphinx-rtd-theme==0.5.2
 sphinxcontrib-applehelp==1.0.2

diff --git a/setup.py b/setup.py
@@ -14,6 +14,7 @@
     "sphinx-autobuild",
     "sphinx-rtd-theme",
     "sphinx-markdown-tables",
+    "sphinx-copybutton",
 ]
 # Packages required for formatting code & running tests.
 extras["test"] = [

diff --git a/textattack/attack_args.py b/textattack/attack_args.py
@@ -108,11 +108,18 @@
 
 
 GOAL_FUNCTION_CLASS_NAMES = {
-    "input-reduction": "textattack.goal_functions.InputReduction",
-    "minimize-bleu": "textattack.goal_functions.MinimizeBleu",
-    "non-overlapping-output": "textattack.goal_functions.NonOverlappingOutput",
-    "targeted-classification": "textattack.goal_functions.TargetedClassification",
-    "untargeted-classification": "textattack.goal_functions.UntargetedClassification",
+    #
+    # Classification goal functions
+    #
+    "targeted-classification": "textattack.goal_functions.classification.TargetedClassification",
+    "untargeted-classification": "textattack.goal_functions.classification.UntargetedClassification",
+    "input-reduction": "textattack.goal_functions.classification.InputReduction",
+    #
+    # Text goal functions
+    #
+    "minimize-bleu": "textattack.goal_functions.text.MinimizeBleu",
+    "non-overlapping-output": "textattack.goal_functions.text.NonOverlappingOutput",
+    "text-to-text": "textattack.goal_functions.text.TextToTextGoalFunction",
 }
 
 

diff --git a/textattack/attack_recipes/input_reduction_feng_2018.py b/textattack/attack_recipes/input_reduction_feng_2018.py
@@ -22,7 +22,7 @@ class InputReductionFeng2018(AttackRecipe):
 
     Pathologies of Neural Models Make Interpretations Difficult.
 
-    ArXiv, abs/1804.07781.
+    https://arxiv.org/abs/1804.07781
     """
 
     @staticmethod

diff --git a/textattack/attack_recipes/textbugger_li_2018.py b/textattack/attack_recipes/textbugger_li_2018.py
@@ -32,7 +32,7 @@ class TextBuggerLi2018(AttackRecipe):
 
     TextBugger: Generating Adversarial Text Against Real-world Applications.
 
-    ArXiv, abs/1812.05271.
+    https://arxiv.org/abs/1812.05271
     """
 
     @staticmethod

diff --git a/textattack/constraints/__init__.py b/textattack/constraints/__init__.py
@@ -1,6 +1,6 @@
 """.. _constraint:
 
-Constraint Package
+Constraints
 ===================
 
 Constraints determine whether a given transformation is valid. Since transformations do not perfectly preserve semantics semantics or grammaticality, constraints can increase the likelihood that the resulting transformation preserves these qualities. All constraints are subclasses of the ``Constraint`` abstract class, and must implement at least one of ``__call__`` or ``call_many``.

diff --git a/textattack/goal_functions/__init__.py b/textattack/goal_functions/__init__.py
@@ -1,7 +1,9 @@
-"""
+""".. _goal_functions:
 
-goal_functions package determine if an attack has been successful.
+Goal Functions
 ==================================================================
+
+Goal Functions determine if an attack has been successful.
 """
 
 from .goal_function import GoalFunction

diff --git a/textattack/goal_functions/classification/input_reduction.py b/textattack/goal_functions/classification/input_reduction.py
@@ -13,8 +13,8 @@ class InputReduction(ClassificationGoalFunction):
     maintaining the same predicted label.
 
     From Feng, Wallace, Grissom, Iyyer, Rodriguez, Boyd-Graber. (2018).
-    Pathologies of Neural Models Make Interpretations Difficult. ArXiv,
-    abs/1804.07781.
+    Pathologies of Neural Models Make Interpretations Difficult.
+    https://arxiv.org/abs/1804.07781
     """
 
     def __init__(self, *args, target_num_words=1, **kwargs):

diff --git a/textattack/metrics/attack_metrics/attack_queries.py b/textattack/metrics/attack_metrics/attack_queries.py
@@ -16,7 +16,7 @@ def __init__(self):
         self.all_metrics = {}
 
     def calculate(self, results):
-        """Calculates all metrics related to number of queries in an attack
+        """Calculates all metrics related to number of queries in an attack.
 
         Args:
             results (``AttackResult`` objects):

diff --git a/textattack/metrics/attack_metrics/attack_success_rate.py b/textattack/metrics/attack_metrics/attack_success_rate.py
@@ -18,7 +18,8 @@ def __init__(self):
         self.all_metrics = {}
 
     def calculate(self, results):
-        """Calculates all metrics related to number of succesful, failed and skipped results in an attack
+        """Calculates all metrics related to number of succesful, failed and
+        skipped results in an attack.
 
         Args:
             results (``AttackResult`` objects):

diff --git a/textattack/metrics/attack_metrics/words_perturbed.py b/textattack/metrics/attack_metrics/words_perturbed.py
@@ -20,7 +20,7 @@ def __init__(self):
         self.all_metrics = {}
 
     def calculate(self, results):
-        """Calculates all metrics related to perturbed words in an attack
+        """Calculates all metrics related to perturbed words in an attack.
 
         Args:
             results (``AttackResult`` objects):

diff --git a/textattack/metrics/metric.py b/textattack/metrics/metric.py
@@ -13,8 +13,7 @@ class Metric(ABC):
     @abstractmethod
     def __init__(self, **kwargs):
         """Creates pre-built :class:`~textattack.Metric` that correspond to
-        evaluation metrics for adversarial examples.
-        """
+        evaluation metrics for adversarial examples."""
         raise NotImplementedError()
 
     @abstractmethod

diff --git a/textattack/metrics/quality_metrics/perplexity.py b/textattack/metrics/quality_metrics/perplexity.py
@@ -39,7 +39,8 @@ def __init__(self, model_name="gpt2"):
         self.stride = 512
 
     def calculate(self, results):
-        """Calculates average Perplexity on all successfull attacks using a pre-trained small GPT-2 model
+        """Calculates average Perplexity on all successfull attacks using a
+        pre-trained small GPT-2 model.
 
         Args:
             results (``AttackResult`` objects):

diff --git a/textattack/metrics/quality_metrics/use.py b/textattack/metrics/quality_metrics/use.py
@@ -20,7 +20,7 @@ def __init__(self, **kwargs):
         self.all_metrics = {}
 
     def calculate(self, results):
-        """Calculates average USE similarity on all successfull attacks
+        """Calculates average USE similarity on all successfull attacks.
 
         Args:
             results (``AttackResult`` objects):

diff --git a/textattack/search_methods/__init__.py b/textattack/search_methods/__init__.py
@@ -1,6 +1,6 @@
 """.. _search_methods:
 
-search_methods package
+Search Methods
 ========================
 
 Search methods explore the transformation space in an attempt to find a successful attack as determined by a :ref:`Goal Functions <goal_function>` and list of :ref:`Constraints <constraint>`

diff --git a/textattack/search_methods/beam_search.py b/textattack/search_methods/beam_search.py
@@ -16,7 +16,7 @@ class BeamSearch(SearchMethod):
 
     Args:
         goal_function: A function for determining how well a perturbation is doing at achieving the attack's goal.
-        transformation (Transformation): The type of transformation.
+        transformation: The type of transformation.
         beam_width (int): the number of candidates to retain at each step
     """
 

diff --git a/textattack/transformations/__init__.py b/textattack/transformations/__init__.py
@@ -7,6 +7,7 @@
 """
 from .transformation import Transformation
 
+from .sentence_transformations import *
 from .word_swaps import *
 from .word_insertions import *
 from .word_merges import *

diff --git a/textattack/transformations/word_innerswap_random.py b/textattack/transformations/word_innerswap_random.py
@@ -1,5 +1,5 @@
 """
-Word Swap by swapping the order of words
+Word Swap Transformation by swapping the order of words
 ==========================================================
 """