trim trailing whitespace

.github/ISSUE_TEMPLATE/a--bug-performance-issue.md

@@ -13,9 +13,9 @@ labels: bugfix
 
 **Describe the expected behavior**
 
-**Code to reproduce the issue**    
+**Code to reproduce the issue**
 <!-- Provide a reproducible test case that is the bare minimum necessary to generate the problem. -->
 
-**Other info / logs**    
-<!-- Include any logs or source code that would be helpful to diagnose the problem. 
+**Other info / logs**
+<!-- Include any logs or source code that would be helpful to diagnose the problem.
 If including tracebacks, please include the full traceback. Large logs and files should be attached. -->

.readthedocs.yml

@@ -13,7 +13,7 @@ sphinx:
 #  configuration: mkdocs.yml
 
 # Optionally build your docs in additional formats such as PDF and ePub
-formats: 
+formats:
     - htmlzip
 
 # Optionally set the version of Python and requirements required to build your docs

CONTRIBUTING.md

@@ -10,7 +10,7 @@ Thank you for contributing to numpy-ml!
 2. For pull requests, please make sure all commits are [*atomic*](https://en.wikipedia.org/wiki/Atomic_commit) (i.e., one feature per commit)
 3. If you're submitting a new model / feature / module, **please include proper documentation and unit tests.**
     - See the `test.py` file in one of the existing modules for examples of unit tests.
-    - Documentation is loosely based on the [NumPy docstring style](https://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_numpy.html). When in doubt, refer to existing examples 
+    - Documentation is loosely based on the [NumPy docstring style](https://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_numpy.html). When in doubt, refer to existing examples
 4. Please format your code using the [black](https://github.com/python/black) defaults. You can use this [online formatter](https://black.now.sh/).
 
 ### Specific guidelines

README.md

@@ -1,6 +1,6 @@
 # numpy-ml
 Ever wish you had an inefficient but somewhat legible collection of machine
-learning algorithms implemented exclusively in numpy? No? 
+learning algorithms implemented exclusively in numpy? No?
 
 ## Documentation
 To see all of the available models, take a look at the [project documentation](https://numpy-ml.readthedocs.io/) or see [here](https://github.com/ddbourgin/numpy-ml/blob/master/numpy_ml/README.md).
@@ -14,4 +14,4 @@ library](https://docs.python.org/3/library/) and [NumPy](https://www.numpy.org/)
 [SciPy library](https://scipy.github.io/devdocs/) is also permitted under special
 circumstances ;)
 
-See full contributing guidelines [here](./CONTRIBUTING.md). 
+See full contributing guidelines [here](./CONTRIBUTING.md).

docs/index.rst

@@ -1,7 +1,7 @@
 Welcome to numpy-ml
 ===================
 `numpy-ml`_ is a growing collection of machine learning models, algorithms, and
-tools written exclusively in `NumPy`_ and the Python `standard library`_. 
+tools written exclusively in `NumPy`_ and the Python `standard library`_.
 
 The purpose of the project is to provide reference implementations of common
 machine learning components for rapid prototyping and experimentation. With
@@ -15,7 +15,7 @@ that in mind, don't just read the docs -- read the source!
 
     We're working to expand our coverage. During this time there are likely to
     be typos, bugs, and poorly-worded sections. If you encounter any of the
-    above, please file an `issue`_ or submit a `pull request`_! 
+    above, please file an `issue`_ or submit a `pull request`_!
 
 .. _issue: https://github.com/ddbourgin/numpy-ml/issues
 .. _pull request: https://github.com/ddbourgin/numpy-ml/pulls
@@ -49,8 +49,8 @@ that in mind, don't just read the docs -- read the source!
    numpy_ml.utils
 
 ##########
-Disclaimer	
-##########	
-	
+Disclaimer
+##########
+
 This software is provided as-is: there are no guarantees that it fits your
 purposes or that it is bug-free. Use it at your own risk!

docs/numpy_ml.gmm.rst

@@ -16,7 +16,7 @@ unsupervised clustering.
 A GMM assumes that:
 
     1. The observed data are generated from a `mixture distribution`_, `P`,
-       made up of `K` mixture components. 
+       made up of `K` mixture components.
 
     2. Each mixture component is a multivariate Gaussian with its own mean
        :math:`\mu`, covariance matrix, :math:`\Sigma`, and mixture weight,
@@ -35,15 +35,15 @@ The parameters of a GMM model are:
       \Sigma_k, \pi_k \}`.
 
 Under a GMM, the joint probability of a sequence of cluster assignments `Z` and an observed
-dataset :math:`X = \{x_1, \ldots, x_N \}`, is: 
+dataset :math:`X = \{x_1, \ldots, x_N \}`, is:
 
 .. math::
-    p(Z, X \mid \theta) = 
-        \prod_{i=1}^N p(z_i, x_i \mid \theta) = 
-            \prod_{i=1}^N \prod_{k=1}^K 
+    p(Z, X \mid \theta) =
+        \prod_{i=1}^N p(z_i, x_i \mid \theta) =
+            \prod_{i=1}^N \prod_{k=1}^K
                 [\mathcal{N}(x_i \mid \mu_k, \Sigma_k) \pi_k ]^{\mathbb{1}_{[z_{i} = k]}}
 
-where 
+where
 
     - :math:`\theta` is the set of GMM parameters: :math:`\theta = \{ \mu_1,
       \Sigma_1, \pi_i, \ldots, \mu_k, \Sigma_k, \pi_k \}`.
@@ -56,7 +56,7 @@ where
       and 0 otherwise.
 
 As with other latent-variable models, we use the `expectation-maximization (EM)
-algorithm`_ to learn the GMM parameters. 
+algorithm`_ to learn the GMM parameters.
 
 .. _expectation-maximization (EM) algorithm : https://en.wikipedia.org/wiki/Expectation%E2%80%93maximization_algorithm
 
@@ -65,7 +65,7 @@ algorithm`_ to learn the GMM parameters.
 - :class:`~numpy_ml.gmm.GMM`
 
 **References**
-    
+
 .. [1] Bilmes, J. A. (1998). "A gentle tutorial of the EM algorithm and its
    application to parameter estimation for Gaussian mixture and hidden
    Markov models" *International Computer Science Institute, 4(510)*

docs/numpy_ml.lda.rst

@@ -1,9 +1,9 @@
 ###########################
-Latent Dirichlet allocation 
+Latent Dirichlet allocation
 ###########################
 
 `Latent Dirichlet allocation`_ (LDA, commonly known as a topic model) is a
-generative model for `bags of words`_.  
+generative model for `bags of words`_.
 
 .. _`Latent Dirichlet allocation` : https://en.wikipedia.org/wiki/Latent_Dirichlet_allocation
 .. _bags of words : https://en.wikipedia.org/wiki/Bag-of-words_model
@@ -47,7 +47,7 @@ The smoothed/fully-Bayesian LDA model [2]_ adds an additional `Dirichlet`_ prior
 
 To generate a document with the smoothed LDA model, we:
 
-    1. Sample the parameters for the distribution over topics, 
+    1. Sample the parameters for the distribution over topics,
        :math:`\theta \sim \text{Dir}(\alpha)`.
 
     2. Sample a topic, :math:`z \sim \text{Cat}(\theta)`.
@@ -63,11 +63,11 @@ The joint distribution over words, topics, :math:`\theta`, and :math:`\phi`
 under the smoothed LDA model is:
 
 .. math::
-    
-    P(w, z, \phi, \theta \mid \alpha, \beta) = \left( \prod_{t=1}^T \text{Dir}(\phi^{(t)}; \beta) \right) \prod_{d=1}^D \text{Dir}(\theta^{(d)}; \alpha) \prod_{n=1}^{N_d} P(z_n \mid \theta^{(d)}) P(w_n \mid \phi^{(z_n)}) 
+
+    P(w, z, \phi, \theta \mid \alpha, \beta) = \left( \prod_{t=1}^T \text{Dir}(\phi^{(t)}; \beta) \right) \prod_{d=1}^D \text{Dir}(\theta^{(d)}; \alpha) \prod_{n=1}^{N_d} P(z_n \mid \theta^{(d)}) P(w_n \mid \phi^{(z_n)})
 
 The parameters of the LDA model can be learned using `variational expectation
-maximization`_ or Markov chain Monte Carlo (e.g., `collapsed Gibbs sampling`_). 
+maximization`_ or Markov chain Monte Carlo (e.g., `collapsed Gibbs sampling`_).
 
 .. _`variational expectation maximization`: https://en.wikipedia.org/wiki/Variational_Bayesian_methods
 .. _`collapsed Gibbs sampling`: https://en.wikipedia.org/wiki/Gibbs_sampling#Collapsed_Gibbs_sampler

docs/numpy_ml.linear_models.rst

@@ -1,4 +1,4 @@
-Linear models 
+Linear models
 #############
 
 The linear models module contains several popular instances of the generalized linear model (GLM).
@@ -230,7 +230,7 @@ We can also compute a closed-form solution for the posterior predictive distribu
 .. math::
 
     y^* \mid X^*, X, Y \sim \mathcal{N}(X^* \mu_b, \ X^* \text{cov}_b X^{* \top} + I)
-    
+
 **Models**
 
 - :class:`~numpy_ml.linear_models.BayesianLinearRegressionUnknownVariance`

docs/numpy_ml.neural_nets.initializers.rst

@@ -1,4 +1,4 @@
-Initializers 
+Initializers
 =============
 
 ``ActivationInitializer``

docs/numpy_ml.neural_nets.layers.rst

@@ -1,4 +1,4 @@
-Layers 
+Layers
 ======
 
 ``LayerBase``

docs/numpy_ml.neural_nets.losses.rst

@@ -1,4 +1,4 @@
-Loss functions 
+Loss functions
 ==============
 
 ``CrossEntropy``

docs/numpy_ml.neural_nets.models.rst

@@ -1,4 +1,4 @@
-Full networks 
+Full networks
 ==============
 
 ``WGAN_GP``

docs/numpy_ml.neural_nets.schedulers.rst

@@ -1,4 +1,4 @@
-Learning rate schedulers 
+Learning rate schedulers
 =========================
 
 ``ConstantScheduler``

docs/numpy_ml.neural_nets.utils.rst

@@ -1,4 +1,4 @@
-Utilities 
+Utilities
 ==========
 
 ``minibatch``

docs/numpy_ml.neural_nets.wrappers.rst

@@ -1,4 +1,4 @@
-Wrappers 
+Wrappers
 =========
 
 ``WrapperBase``

docs/numpy_ml.ngram.rst

@@ -13,7 +13,7 @@ denote the `(j - i)`-gram :math:`(w_{i}, w_{i+1}, \ldots, w_{j})`.
    <h3>Laplace Smoothing</h3>
 
 `Laplace smoothing`_ is the assumption that each `n`-gram in a corpus occurs
-exactly one more time than it actually does. 
+exactly one more time than it actually does.
 
 .. math::
 

docs/numpy_ml.nonparametric.rst

@@ -1,4 +1,4 @@
-Nonparametric models 
+Nonparametric models
 ####################
 
 .. raw:: html

docs/numpy_ml.preprocessing.dsp.rst

@@ -1,4 +1,4 @@
-Digital signal processing 
+Digital signal processing
 #########################
 
 ``DCT``

docs/numpy_ml.preprocessing.rst

@@ -1,4 +1,4 @@
-Preprocessing 
+Preprocessing
 #############
 
 .. toctree::

docs/numpy_ml.rl_models.agents.rst

@@ -1,4 +1,4 @@
-Agents 
+Agents
 ======
 
 ``CrossEntropyAgent``

docs/numpy_ml.rl_models.rl_utils.rst

@@ -1,4 +1,4 @@
-Utilities 
+Utilities
 =========
 
 .. automodule:: numpy_ml.rl_models.rl_utils

docs/numpy_ml.rl_models.rst

@@ -1,4 +1,4 @@
-Reinforcement learning 
+Reinforcement learning
 ######################
 
 .. toctree::

docs/numpy_ml.trees.losses.rst

@@ -6,4 +6,4 @@ Losses (``trees.losses``)
 	:members:
 	:undoc-members:
 	:inherited-members:
-	:show-inheritance:	
+	:show-inheritance:

docs/numpy_ml.trees.rst

@@ -1,4 +1,4 @@
-Tree-based models 
+Tree-based models
 #################
 .. raw:: html
 
@@ -35,24 +35,24 @@ in an inhomogeneity or "impurity" metric, :math:`\mathcal{L}`. One popular
 metric is the **information entropy**:
 
 .. math::
-    
+
     -\sum_j P_n(\omega_j) \log P_n(\omega_j)
 
 where :math:`P_n(\omega_j)` is the fraction of data points at split `n` that are
 associated with category :math:`\omega_j`. Another useful metric is the **Gini
 impurity**:
 
 .. math::
-    
+
     \sum_{i \neq j} P_n(\omega_i) P_n(\omega_j) = 1 - \sum_{j} P_n(\omega_j)^2
 
 For a binary tree (where each node has only two children), the reduction in
 impurity after a particular split is
 
 .. math::
-    
-    \Delta \mathcal{L} = \mathcal{L}(\text{Parent}) - 
-        P_{left} \mathcal{L}(\text{Left child}) - 
+
+    \Delta \mathcal{L} = \mathcal{L}(\text{Parent}) -
+        P_{left} \mathcal{L}(\text{Left child}) -
             (1 - P_{left})\mathcal{L}(\text{Right child})
 
 where :math:`\mathcal{L}(x)` is the impurity of the dataset at node `x`,
@@ -88,7 +88,7 @@ aggregating. For this approach, each of the `n` learners is a different
 decision tree. In addition to training each decision tree on a different
 bootstrapped dataset, random forests employ a `random subspace`_ approach [5]_:
 each decision tree is trained on a subsample (without replacement) of the full
-collection of dataset features. 
+collection of dataset features.
 
 .. _`Bootstrap aggregating`: https://en.wikipedia.org/wiki/Bootstrap_aggregating
 .. _`random forest`: https://en.wikipedia.org/wiki/Random_forest
@@ -102,7 +102,7 @@ collection of dataset features.
 **References**
 
 .. [2] Breiman, L. (1994). "Bagging predictors". *Technical Report 421.
-   Statistics Department, UC Berkeley*. 
+   Statistics Department, UC Berkeley*.
 .. [3] Ho, T. K. (1995). "Random decision forests". *Proceedings of the Third
    International Conference on Document Analysis and Recognition, 1*: 278-282.
 .. [4] Breiman, L. (2001). "Random forests". *Machine Learning. 45(1)*: 5-32.
@@ -152,7 +152,7 @@ weak learners.
 **References**
 
 .. [6]  Breiman, L. (1997). "Arcing the edge". *Technical Report 486.
-   Statistics Department, UC Berkeley*. 
+   Statistics Department, UC Berkeley*.
 .. [7] Friedman, J. H. (1999). "Greedy function approximation: A gradient
    boosting machine". *IMS 1999 Reitz Lecture*.
 .. [8]  Mason, L., Baxter, J., Bartlett, P. L., Frean, M. (1999). "Boosting

docs/numpy_ml.utils.data_structures.rst

@@ -1,4 +1,4 @@
-Data structures 
+Data structures
 ================
 
 ``BallTree``

docs/numpy_ml.utils.distance_metrics.rst

@@ -1,4 +1,4 @@
-Distance metrics 
+Distance metrics
 ================
 
 Common distance functions.

docs/numpy_ml.utils.kernels.rst

@@ -1,8 +1,8 @@
-Kernels 
+Kernels
 =======
 
 A collection of common kernel / similarity functions. All kernels are
-continuous, bounded, and symmetric real functions which integrate to 1. 
+continuous, bounded, and symmetric real functions which integrate to 1.
 
 ``LinearKernel``
 ----------------

docs/numpy_ml.utils.testing.rst

@@ -1,4 +1,4 @@
-Testing 
+Testing
 -------
 Common helper functions for testing the ML algorithms in the rest of the repo.
 

docs/requirements.txt

@@ -1,8 +1,8 @@
 numpy
-scipy 
+scipy
 
 # all this is for the dang tests
-matplotlib 
+matplotlib
 seaborn
 pandas
 sklearn