Used as an adjective, "pairwise-comparison" takes a hyphen.

README.md

@@ -37,7 +37,7 @@ To install the latest release directly from PyPI, simply type
 
 To get started, you might want to explore one of these notebooks:
 
-- [Introduction using pairwise comparison data](notebooks/intro-pairwise.ipynb)
+- [Introduction using pairwise-comparison data](notebooks/intro-pairwise.ipynb)
 - [Case study: analyzing the GIFGIF dataset](notebooks/gifgif-dataset.ipynb)
 - [Using ChoiceRank to understand traffic on a network](notebooks/choicerank-tutorial.ipynb)
 - [Approximate Bayesian inference using EP](notebooks/ep-example.ipynb)

choix/ep.py

@@ -34,7 +34,7 @@ def ep_pairwise(n_items, data, alpha, model="logit", max_iter=100,
  """Compute a distribution of model parameters using the EP algorithm.
 
  This function computes an approximate Bayesian posterior probability
- distribution over model parameters, given pairwise comparison data (see
+ distribution over model parameters, given pairwise-comparison data (see
  :ref:`data-pairwise`). It uses the expectation propagation algorithm, as
  presented, e.g., in [CG05]_.
 
@@ -43,15 +43,15 @@ def ep_pairwise(n_items, data, alpha, model="logit", max_iter=100,
  Gaussian distribution, described by a mean vector and a covariance matrix.
 
  Two different observation models are available. ``logit`` (default) assumes
- that pairwise comparison outcomes follow from a Bradley-Terry model.
+ that pairwise-comparison outcomes follow from a Bradley-Terry model.
  ``probit`` assumes that the outcomes follow from Thurstone's model.
 
  Parameters
  ----------
  n_items : int
  Number of distinct items.
  data : list of lists
- Pairwise comparison data.
+ Pairwise-comparison data.
  alpha : float
  Inverse variance of the (isotropic) prior.
  model : str, optional

choix/lsr.py

@@ -37,7 +37,7 @@ def lsr_pairwise(n_items, data, alpha=0.0, initial_params=None):
  """Compute the LSR estimate of model parameters.
 
  This function implements the Luce Spectral Ranking inference algorithm
- [MG15]_ for pairwise comparison data (see :ref:`data-pairwise`).
+ [MG15]_ for pairwise-comparison data (see :ref:`data-pairwise`).
 
  The argument ``initial_params`` can be used to iteratively refine an
  existing parameter estimate (see the implementation of
@@ -53,7 +53,7 @@ def lsr_pairwise(n_items, data, alpha=0.0, initial_params=None):
  n_items : int
  Number of distinct items.
  data : list of lists
- Pairwise comparison data.
+ Pairwise-comparison data.
  alpha : float, optional
  Regularization parameter.
  initial_params : array_like, optional
@@ -76,7 +76,7 @@ def ilsr_pairwise(
  """Compute the ML estimate of model parameters using I-LSR.
 
  This function computes the maximum-likelihood (ML) estimate of model
- parameters given pairwise comparison data (see :ref:`data-pairwise`), using
+ parameters given pairwise-comparison data (see :ref:`data-pairwise`), using
  the iterative Luce Spectral Ranking algorithm [MG15]_.
 
  The transition rates of the LSR Markov chain are initialized with
@@ -88,7 +88,7 @@ def ilsr_pairwise(
  n_items : int
  Number of distinct items.
  data : list of lists
- Pairwise comparison data.
+ Pairwise-comparison data.
  alpha : float, optional
  Regularization parameter.
  initial_params : array_like, optional
@@ -213,7 +213,7 @@ def rank_centrality(n_items, data, alpha=0.0):
  n_items : int
  Number of distinct items.
  data : list of lists
- Pairwise comparison data.
+ Pairwise-comparison data.
  alpha : float, optional
  Regularization parameter.
 

choix/mm.py

@@ -48,7 +48,7 @@ def mm_pairwise(
  """Compute the ML estimate of model parameters using the MM algorithm.
 
  This function computes the maximum-likelihood (ML) estimate of model
- parameters given pairwise comparison data (see :ref:`data-pairwise`), using
+ parameters given pairwise-comparison data (see :ref:`data-pairwise`), using
  the minorization-maximization (MM) algorithm [Hun04]_, [CD12]_.
 
  If ``alpha > 0``, the function returns the maximum a-posteriori (MAP)
@@ -60,7 +60,7 @@ def mm_pairwise(
  n_items : int
  Number of distinct items.
  data : list of lists
- Pairwise comparison data.
+ Pairwise-comparison data.
  initial_params : array_like, optional
  Parameters used to initialize the iterative procedure.
  alpha : float, optional

choix/opt.py

@@ -14,11 +14,11 @@ def _safe_exp(x):
 
 class PairwiseFcts:
 
- """Optimization-related methods for pairwise comparison data.
+ """Optimization-related methods for pairwise-comparison data.
  
  This class provides methods to compute the negative log-likelihood (the
  "objective"), its gradient and its Hessian, given model parameters and
- pairwise comparison data.
+ pairwise-comparison data.
  """
 
  def __init__(self, data, penalty):
@@ -128,15 +128,15 @@ def opt_pairwise(n_items, data, penalty=1e-6, method="Newton-CG",
  """Compute the ML estimate of model parameters using ``scipy.optimize``.
 
  This function computes the (penalized) maximum-likelihood estimate of model
- parameters given pairwise comparison data (see :ref:`data-pairwise`), using
+ parameters given pairwise-comparison data (see :ref:`data-pairwise`), using
  optimizers provided by the ``scipy.optimize`` module.
 
  Parameters
  ----------
  n_items : int
  Number of distinct items.
  data : list of lists
- Pairwise comparison data.
+ Pairwise-comparison data.
  penalty : float, optional
  Regularization strength.
  method : str, optional

choix/utils.py

@@ -279,7 +279,7 @@ def generate_pairwise(params, n_comparisons=10):
  Returns
  -------
  data : list of (int, int)
- Pairwise comparison samples (see :ref:`data-pairwise`).
+ Pairwise-comparison samples (see :ref:`data-pairwise`).
  """
  n = len(params)
  items = tuple(range(n))

notebooks/gifgif-dataset.ipynb

@@ -16,7 +16,7 @@
  "Therefore, the raw data that is collected consists of outcomes of pairwise comparison between pairs of images.\n",
  "*Just the kind of data that `choix` is built for*!\n",
  "\n",
- "In this notebook, we will **use `choix` to try making sense of the raw pairwise comparison data**.\n",
+ "In this notebook, we will **use `choix` to try making sense of the raw pairwise-comparison data**.\n",
  "In particular, we would like to embed the images on a scale (for a given emotion).\n",
  "\n",
  "### Dataset\n",
@@ -95,7 +95,7 @@
  "## Processing the raw data\n",
  "\n",
  "First, we need to transform the raw dataset into a format that `choix` can process.\n",
- "Remember that `choix` encodes pairwise comparison outcomes as tuples `(i, j)` (meaning \"$i$ won over $j$\"), and that items are assumed to be numbered by consecutive integers.\n",
+ "Remember that `choix` encodes pairwise-comparison outcomes as tuples `(i, j)` (meaning \"$i$ won over $j$\"), and that items are assumed to be numbered by consecutive integers.\n",
  "\n",
  "We begin by mapping all distinct images that appear in the dataset to consecutive integers."
  ]

notebooks/intro-pairwise.ipynb

@@ -4,10 +4,10 @@
  "cell_type": "markdown",
  "metadata": {},
  "source": [
- "# Introduction using pairwise comparison data\n",
+ "# Introduction using pairwise-comparison data\n",
  "\n",
  "This notebook provides a gentle introduction to the `choix` library.\n",
- "We consider the case of pairwise comparison outcomes between items from some set."
+ "We consider the case of pairwise-comparison outcomes between items from some set."
  ]
  },
  {