Add codespell support (config, workflow to detect/not fix) and make it fix few typos

.github/workflows/codespell.yml

@@ -0,0 +1,25 @@
+# Codespell configuration is within setup.cfg
+---
+name: Codespell
+
+on:
+  push:
+    branches: [master]
+  pull_request:
+    branches: [master]
+
+permissions:
+  contents: read
+
+jobs:
+  codespell:
+    name: Check for spelling errors
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Annotate locations with typos
+        uses: codespell-project/codespell-problem-matcher@v1
+      - name: Codespell
+        uses: codespell-project/actions-codespell@v2

doc/developers/style_guide.rst

@@ -20,7 +20,7 @@ Coding style
 ============
 
 pulse2percept uses the standard Python `PEP8`_ style guide to ensure
-readibility and consistency of the code base.
+readability and consistency of the code base.
 
 .. note::
 
@@ -44,7 +44,7 @@ Imports
 -------
 
 pulse2percept recommends using the following package shorthands to increase
-consistency and readibility across the library:
+consistency and readability across the library:
 
 .. code-block:: python
 
@@ -92,7 +92,7 @@ The recommended style is to break a line after binary operators:
               - student_loan_interest)
 
 Yes, we are aware of the ongoing `W503 vs W504 discussion <https://www.python.org/dev/peps/pep-0008/#should-a-line-break-before-or-after-a-binary-operator>`_.
-W504 would improve readibility.
+W504 would improve readability.
 However, for now, we want to ensure consistency with the existing code.
 
 Class layout (slots, properties)

doc/install.rst

@@ -290,7 +290,7 @@ You can check the installation location:
 
    pip show pulse2percept
 
-Then add the specificed location to ``$PATH``; see `PATH on Windows`_, 
+Then add the specified location to ``$PATH``; see `PATH on Windows`_, 
 `PATH on macOS`_, `PATH on Linux`_.
 
 .. _PATH on Windows: https://helpdeskgeek.com/windows-10/add-windows-path-environment-variable/

doc/topics/cortical.rst

@@ -426,7 +426,7 @@ For Developers
 --------------
 
 In this section we will discuss some of the changes made under the hood
-accomadate cortical features, as well as some important notes for developers
+accommodate cortical features, as well as some important notes for developers
 to keep in mind.
 
 Units
@@ -449,7 +449,7 @@ the model plotting section.
 In order to create your own visual field map, you must create a subclass of
 :py:class:`~pulse2percept.topography.CorticalMap`, and implement the `dva_to_v1`
 method.  In addition, if your map also maps to v2 and/or v3, you must also
-implement the `dva_to_v2` and/or `dva_to_v3` methods. Optinally, you can also
+implement the `dva_to_v2` and/or `dva_to_v3` methods. Optionally, you can also
 implement `v1_to_dva`, `v2_to_dva`, and/or `v3_to_dva` methods.
 
 For example, if you wanted to create a map that mapped `(x, y)` in dva to

doc/users/release_notes.rst

@@ -174,7 +174,7 @@ API changes
 Backward-incompatible changes
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-*  Implants rotation angles are now speicified in degrees, not radians (:pull:`357`)
+*  Implants rotation angles are now specified in degrees, not radians (:pull:`357`)
 *  pulse2percept now requires Matplotlib 3.0.2 or newer (:pull:`223`)
 *  FFMPEG and scikit-video dependencies have been removed (:pull:`196`)
 *  ``TimeSeries`` has been removed. Please use

examples/models/plot_granley2021_biphasic.py

@@ -305,7 +305,7 @@ def size_modulation(freq, amp, pdur):
 # its faster, stripped version `_predict_spatial_batched`. This 
 # function is only intended to be used if you are repeatedly simulating batches of percepts. 
 # Since jax compiles each function the first time it is used, using this function only once
-# for a singular group of stimuli will be noticably slower than repeatedly applying 
+# for a singular group of stimuli will be noticeably slower than repeatedly applying 
 # `predict_percept`. However, splitting a very large set of stimuli into smaller batches and 
 # using `predict_percept_batched` will be significantly faster than `predict_percept` on each
 # individual stimuli.

examples/models/plot_nanduri2012.py

@@ -72,7 +72,7 @@
 #
 # Interacting with a model always involves three steps:
 #
-# 1.  **Initalize** the model by passing the desired parameters.
+# 1.  **Initialize** the model by passing the desired parameters.
 # 2.  **Build** the model to perform (sometimes expensive) one-time setup
 #     computations.
 # 3.  **Predict** the percept for a given stimulus.

examples/models/plot_visual_field_maps.py

@@ -37,7 +37,7 @@
 :py:class:`~pulse2percept.topography.CorticalMap`.
 This means that all retinal visual field maps have to specify a ``dva_to_ret`` method, 
 which transforms visual field coordinates into retinal coordinates, and all cortical 
-visual field maps have to specify atleast a ``dva_to_v1`` method, which transforms visual 
+visual field maps have to specify at least a ``dva_to_v1`` method, which transforms visual 
 field coordinates into cortical V1 coordinates. Cortical models map also specify ``dva_to_v2``
 and ``dva_to_v3`` methods, which transform visual field coordinates into cortical V2 and V3.
 Most visual field maps also specify the inverse transform, e.g. ``ret_to_dva`` or ``v1_to_dva``.

examples/stimuli/plot_psychophysics.py

@@ -6,7 +6,7 @@
 
 *This example shows how to use drifting psychophysics-based stimuli for a retinal implant.*
 
-Along with images, videos, and oter built-in stimuli, pulse2percept supports
+Along with images, videos, and other built-in stimuli, pulse2percept supports
 generating :py:class:`~pulse2percept.stimuli.GratingStimulus` and :py:class:`~pulse2percept.stimuli.BarStimulus` as stimuli
 that can be passed as percepts to implants.
 

pulse2percept/implants/cortex/tests/test_neuralink.py

@@ -31,7 +31,7 @@ def test_LinearEdgeThread():
     npt.assert_almost_equal(thread.y, 0)
     npt.assert_almost_equal(thread.z, 0)
 
-    # elecs arent actually at this spot, but are on the edge, a few microns off
+    # elecs aren't actually at this spot, but are on the edge, a few microns off
     zs = []
     for e in thread.electrode_objects:
         npt.assert_almost_equal(e.x, thread.r + 7 // 2)

pulse2percept/model_selection/optimizers.py

@@ -138,7 +138,7 @@ def fit(self, X, y=None, fit_params=None):
         self._optimize(X, y, fit_params=fit_params)
         if self.verbose:
             print(f'Best score: {self._best_score}, Best params: {self._best_params}')
-        # Use the parameter values found during optimiziation to set the
+        # Use the parameter values found during optimization to set the
         # estimator one more time, then fit it:
         self.estimator.set_params(**self._best_params)
         if fit_params is None:
@@ -262,8 +262,8 @@ class FunctionMinimizer(BaseOptimizer):
     estimator : ``sklearn.base.estimator``
         A scikit-learn estimator, such as a classifier or regressor, that
         contains ``fit`` and ``predict`` methods.
-    serch_params : dict of tupels (lower bound, upper bound)
-        Dictionary of tupels containing the lower and upper bound of values
+    serch_params : dict of tuples (lower bound, upper bound)
+        Dictionary of tuples containing the lower and upper bound of values
         for each search parameter.
     search_params_init : dict of floats, optional
         Initial values of all search parameters. If None, initialize to
@@ -324,7 +324,7 @@ def _optimize(self, X, y, fit_params=None):
         res = minimize(self._calc_error, init, args=(X, y, fit_params),
                        bounds=bounds, tol=self.tol, options=self.options)
         if not res['success']:
-            print('Optimization unsucessful:')
+            print('Optimization unsuccessful:')
             print(res)
 
         # Pair values of best params with their names to build a dict
@@ -347,8 +347,8 @@ class ParticleSwarmOptimizer(BaseOptimizer):
     estimator : ``sklearn.base.estimator``
         A scikit-learn estimator, such as a classifier or regressor, that
         contains ``fit`` and ``predict`` methods.
-    serch_params : dict of tupels (lower bound, upper bound)
-        Dictionary of tupels containing the lower and upper bound of values
+    serch_params : dict of tuples (lower bound, upper bound)
+        Dictionary of tuples containing the lower and upper bound of values
         for each search parameter.
     swarm_size : int, optional, default: 10 * number of search params
         The number of particles in the swarm.

pulse2percept/models/_nanduri2012.pyx

@@ -38,7 +38,7 @@ cpdef spatial_fast(const float32[:, ::1] stim,
         An array of x or y coordinates at which to calculate the spatial
         response (microns)
     atten_a : float32
-        Nominator of the attentuation function (Eq.2 in the paper)
+        Nominator of the attenuation function (Eq.2 in the paper)
     atten_n : float32
         Exponent of the attenuation function's denominator (Eq.2 in the paper)
     thresh_percept : float32
@@ -91,7 +91,7 @@ cpdef spatial_fast(const float32[:, ::1] stim,
                     denom = atten_a + c_pow(zel[idx_el], atten_n)
                 else:
                     # Away from the electrode surface, calculate the distance
-                    # to the electrode egde (`d2e`):
+                    # to the electrode edge (`d2e`):
                     d2e = (c_pow(c_sqrt(d2c) - rel[idx_el], 2) +
                            c_pow(zel[idx_el], 2))
                     denom = atten_a + c_pow(c_sqrt(d2e), atten_n)

pulse2percept/models/base.py

@@ -757,7 +757,7 @@ def __getattr__(self, attr):
             *  returns the attribute if found.
             *  if the attribute exists in both spatial / temporal model,
                returns a dictionary ``{'spatial': attr, 'temporal': attr}``.
-            *  if the attribtue is not found, raises an AttributeError.
+            *  if the attribute is not found, raises an AttributeError.
 
         """
         # Check the spatial/temporal model:

pulse2percept/models/beyeler2019.py

@@ -474,7 +474,7 @@ def find_closest_axon(self, bundles, xret=None, yret=None,
             Note that each bundle will most likely have a different N
         xret, yret : scalar or list of scalars
             The x,y location on the retina (in microns, where the fovea is the
-            origin) for which to find the closests axon.
+            origin) for which to find the closest axon.
         return_index : bool, optional
             If True, the function will also return the index into ``bundles``
             that represents the closest axon

pulse2percept/models/cortex/tests/test_base.py

@@ -75,7 +75,7 @@ def test_predict_spatial(ModelClass, regions):
     npt.assert_equal(np.all(percept.data[:, half+1:] == 0), True)
     npt.assert_equal(np.all(percept.data[:, :half] != 0), True)
 
-    # implant only in v1, shouldnt change with v2/v3
+    # implant only in v1, shouldn't change with v2/v3
     vfmap = Polimeni2006Map(k=15, a=0.5, b=90)
     model = ModelClass(xrange=(-5, 0), yrange=(-3, 3), xystep=0.1, rho=400, vfmap=vfmap).build()
     elecs = [79, 49, 19, 80, 50, 20, 90, 61, 31, 2, 72, 42, 12, 83, 53, 23, 93, 64, 34, 5, 75, 45, 15, 86, 56, 26, 96, 67, 37, 8, 68, 38]
@@ -171,7 +171,7 @@ def test_plot(ModelClass):
 
 
 def test_poli_nlink():
-    # make sure that the polimeni map and neuralink work togther with scoreboard
+    # make sure that the polimeni map and neuralink work together with scoreboard
     # since this is an odd combo of 2d map and 3d implant
     model = ScoreboardModel(rho=800, xystep=.5).build()
     npt.assert_equal(model.grid.v1.z is None, True)

pulse2percept/models/granley2021.py

@@ -347,7 +347,7 @@ def __setattr__(self, name, value):
         # try to set it ourselves, but can't use get_attr
         try:
             self.__getattribute__(name)
-            # if we get here, we have the attribute, not (neccesarily) an effects model
+            # if we get here, we have the attribute, not (necessarily) an effects model
             super().__setattr__(name, value)
             found = True
         except AttributeError:
@@ -571,7 +571,7 @@ def predict_one(e_params):
     def predict_percept(self, implant, t_percept=None):
         """ Predicts the spatial response
         Override base predict percept to have desired timesteps and 
-        remove unneccesary computation
+        remove unnecessary computation
 
         Parameters
         ----------

pulse2percept/models/nanduri2012.py

@@ -35,7 +35,7 @@ class Nanduri2012Spatial(SpatialModel):
     Parameters
     ----------
     atten_a : float, optional
-        Nominator of the attentuation function
+        Nominator of the attenuation function
     atten_n : float32, optional
         Exponent of the attenuation function's denominator
     vfmap : :py:class:`~pulse2percept.topography.VisualFieldMap`, optional
@@ -198,7 +198,7 @@ class Nanduri2012Model(Model):
     Parameters
     ----------
     atten_a : float, optional
-        Nominator of the attentuation function (Eq.2 in the paper)
+        Nominator of the attenuation function (Eq.2 in the paper)
     atten_n : float32, optional
         Exponent of the attenuation function's denominator (Eq.2 in the paper)
     dt : float, optional

pulse2percept/models/tests/test_granley2021.py

@@ -314,7 +314,7 @@ def test_biphasicAxonMapModel(engine):
     for atr in ['a' + str(i) for i in range(0, 10)]:
         npt.assert_equal(hasattr(model, atr), True)
     model.a0 = 5
-    # Should propogate to size and bright model
+    # Should propagate to size and bright model
     # But should not be a member of streak or spatial
     npt.assert_equal(model.spatial.size_model.a0, 5)
     npt.assert_equal(model.spatial.bright_model.a0, 5)
@@ -337,11 +337,11 @@ def test_biphasicAxonMapModel(engine):
     npt.assert_equal(model.rho, 432)
     npt.assert_equal(model.spatial.size_model.rho, 432)
 
-    # If parameter is not an effects model param, it cant be set
+    # If parameter is not an effects model param, it can't be set
     with pytest.raises(FreezeError):
         model.invalid_param = 5
 
-    # Custom parameters also propogate to effects models
+    # Custom parameters also propagate to effects models
     model = BiphasicAxonMapModel(engine=engine)
 
     class TestSizeModel():
@@ -361,7 +361,7 @@ def __call__(self, freq, amp, pdur):
     class TestInitClassGood():
         def __init__(self):
             self.model = BiphasicAxonMapModel()
-            # This shouldnt raise an error
+            # This shouldn't raise an error
             self.model.a0
 
     class TestInitClassBad():

pulse2percept/stimuli/base.py

@@ -118,7 +118,7 @@ class Stimulus(PrettyPrint):
            The number of time points provided must match the number of time
            points extracted from the source type (i.e., M).
            Stimuli created from scalars or 1-D NumPy arrays will have no time
-           componenet, in which case you cannot provide your own time points.
+           component, in which case you cannot provide your own time points.
 
     metadata : dict, optional
         Additional stimulus metadata can be stored in a dictionary.
@@ -208,7 +208,7 @@ def _from_source(self, source):
             time = None
             electrodes = None
         elif isinstance(source, (list, tuple)):
-            # List or touple with N elements: 1 electrode, N time points
+            # List or tuple with N elements: 1 electrode, N time points
             data = np.array(source, dtype=np.float32).reshape((1, -1))
             time = np.arange(data.shape[-1], dtype=np.float32)
             electrodes = None

pulse2percept/topography/base.py

@@ -23,7 +23,7 @@
 class CoordinateGrid:
         """
         Datatype for storing a grid of coordinates
-        Basically an overriden namedtuple with custom __eq__ method
+        Basically an overridden namedtuple with custom __eq__ method
         """
         x : np.ndarray
         y : np.ndarray

pulse2percept/topography/cortex.py

@@ -119,7 +119,7 @@ def _invert_left_pol(self, theta, radius, inverted = None):
 
     def _invert_left_cart(self, x, y, inverted = None, boundary=0):
         """
-        'Corrects' the mapping by flipping x axis if neccesary, allowing for both
+        'Corrects' the mapping by flipping x axis if necessary, allowing for both
          left and right hemisphere to use the same map.
         """
         # Check if we're reverting from an existing inversion

pulse2percept/topography/tests/test_neuropythy.py

@@ -132,7 +132,7 @@ def test_dva_to_cortex(regions, jitter_boundary):
 def test_Neuralink_from_neuropythy():
     nmap = NeuropythyMap('fsaverage', regions=['v1'], jitter_boundary=False)
     nlink = Neuralink.from_neuropythy(nmap, locs=np.array([[0, 0], [3, 3], [-2, -2]]))
-    # 0, 0 should be nan so it wont make one
+    # 0, 0 should be nan so it won't make one
     npt.assert_equal(len(nlink.implants), 2)
     npt.assert_equal(nlink.implants['A'].x, nmap.dva_to_v1(3, 3, surface='pial')[0])
     npt.assert_equal(nlink.implants['A'].y, nmap.dva_to_v1(3, 3, surface='pial')[1])

setup.cfg

@@ -2,3 +2,10 @@
 universal=1
 [easy_install]
 zip_ok = False
+
+[codespell]
+# Ref: https://github.com/codespell-project/codespell#using-a-config-file
+skip = .git*,*.css
+check-hidden = true
+# ignore-regex = 
+ignore-words-list = numer,theses

setup.py

@@ -183,7 +183,7 @@ def configuration(parent_package='', top_path=None):
 
     config = Configuration(None, parent_package, top_path)
 
-    # Avoid non-useful msg:
+    # Avoid useless msg:
     # "Ignoring attempt to set 'name' (from ... "
     config.set_options(ignore_setup_xxx_py=True,
                        assume_default_configuration=True,