Merge pull request #22 from EthicallyAI/dev

Version 0.0.4
ResponsiblyAI · Jun 3, 2019 · 13dc98d · 13dc98d
2 parents 8af2aa8 + c779693
commit 13dc98d
Show file tree

Hide file tree

Showing 17 changed files with 3,225 additions and 3,609 deletions.
diff --git a/.pycodestyle.ini b/.pycodestyle.ini
@@ -6,4 +6,4 @@
 # E711: comparison to None (used to improve test style)
 # E712: comparison to True (used to improve test style)
 # W503 line break before binary operator
-ignore = E401,E402,E501,E711,E712,W503
+ignore = E401,E402,E501,E711,E712,W503,E127
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -1,6 +1,21 @@
 Revision History
 ================
 
+0.0.4 (2019/06/03)
+------------------
+
+- Word embedding bias
+
+  - Unrestricted ``most_similar``
+
+  - Unrestricted ``generate_analogies``
+
+  - Running specific experiments with ``calc_all_weat``
+
+  - Plotting clustering by classification
+    of biased neutral words
+
+
 0.0.3 (2019/04/10)
 ------------------
 
@@ -15,7 +30,7 @@ Revision History
   - Equalized odds post-processing algorithmic interventions
   - Complete two notebook demos (FICO and COMPAS)
 
-- Words embedding bias
+- Word embedding bias
 
   - Measuring bias with WEAT method
 
@@ -27,16 +42,16 @@ Revision History
 0.0.2 (2018/09/01)
 ------------------
 
-- Words embedding bias
+- Word embedding bias
 
   - Generating analogies along the bias direction
-  - Standard evaluations of words embedding (word pairs and analogies)
+  - Standard evaluations of word embedding (word pairs and analogies)
   - Plotting indirect bias
-  - Scatter plot of bias direction projections between two words embedding
+  - Scatter plot of bias direction projections between two word embedding
   - Improved verbose mode
 
 
 0.0.1 (2018/08/17)
 ------------------
 
--  Gender debiasing for words embedding based on Bolukbasi et al.
+-  Gender debiasing for word embedding based on Bolukbasi et al.
diff --git a/README.rst b/README.rst
@@ -27,7 +27,7 @@ Ethically
     :target: http://docs.ethically.ai/about/license.html
 
 **Toolkit for Auditing and Mitigating Bias and Fairness**
-**of Machine Learning Systems 🔎🤖🔧**
+**of Machine Learning Systems 🔎🤖🧰**
 
 *Ethically* is developed for **practitioners** and **researchers** in mind,
 but also for learners. Therefore, it is compatible with
@@ -51,7 +51,7 @@ Besides, there is a particular focus on **NLP** models.
 
 3. ``ethically.we``
      Metrics and debiasing methods for bias (such as gender and race)
-     in words embedding.
+     in word embedding.
 
 For fairness, *Ethically*'s functionality is aligned with the book
 `Fairness and Machine Learning

diff --git a/docs/demos.rst b/docs/demos.rst
@@ -6,7 +6,7 @@ Demos
 
   notebooks/demo-compas-analysis
   notebooks/demo-fico-analysis
-  notebooks/demo-words-embedding-bias
+  notebooks/demo-word-embedding-bias
 
 The demos can also be accessible as Jupyter notebooks
 through the `GitHub repository

diff --git a/docs/index.rst b/docs/index.rst
@@ -14,7 +14,7 @@ Contents
 
   dataset
   fairness
-  words-embedding-bias
+  word-embedding-bias
   demos
   about/contributing
   about/changelog

diff --git a/...notebooks/demo-words-embedding-bias.ipynb → .../notebooks/demo-word-embedding-bias.ipynb b/...notebooks/demo-words-embedding-bias.ipynb → .../notebooks/demo-word-embedding-bias.ipynb
diff --git a/docs/words-embedding-bias.rst → docs/word-embedding-bias.rst b/docs/words-embedding-bias.rst → docs/word-embedding-bias.rst
@@ -1,9 +1,9 @@
-Words Embedding Bias
-====================
+Word Embedding Bias
+===================
 
 .. automodule:: ethically.we
 
-Refer to the `Words Embedding demo <notebooks/demo-words-embedding-bias.html>`_
+Refer to the `Word Embedding demo <notebooks/demo-word-embedding-bias.html>`_
 for a complete usage example.
 
 Bolukbasi Bias Measure and Debiasing
@@ -18,9 +18,14 @@ WEAT
 .. automodule:: ethically.we.weat
     :members:
 
+Utilities
+---------
 
-Words Embedding Benchmarks
---------------------------
+.. automodule:: ethically.we.utils
+    :members:
+
+Word Embedding Benchmarks
+-------------------------
 
 .. automodule:: ethically.we.benchmark
     :members:
diff --git a/ethically/__init__.py b/ethically/__init__.py
@@ -1,9 +1,9 @@
+# pylint: disable=line-too-long
+
 __project__ = 'ethically'
-__description__ = ('Toolkit for Auditing and Mitigating Bias'
-                   ' and Fairness of Machine Learning Systems'
-                   ' 🔎🤖🔧')
+__description__ = 'Toolkit for Auditing and Mitigating Bias and Fairness of Machine Learning Systems 🔎🤖🧰'
 __url__ = 'http://docs.ethically.ai'
-__version__ = '0.0.3'
+__version__ = '0.0.4'
 __author__ = 'Shlomi Hod'
 __author_email__ = 'shlomi.hod@gmail.com'
 __license__ = 'MIT'