Fix docs and docstrings

docs/user_guide/fit_predict.rst

@@ -58,7 +58,7 @@ The default quantiles can be overwritten at prediction time by specifying a valu
     >>> y_pred.ndim == 1
     True
 
-The output of the `predict` method is an array with one column for each specified quantile or a single column if no quantiles are specified. The order of the output columns corresponds to the order of the quantiles, which can be specified in any order (i.e., they do not need to be monotonically ordered)::
+The output of the `predict` method is an array with one column for each specified quantile or a single column if an individual quantile is specified. The order of the output columns corresponds to the order of the quantiles, which can be specified in any order (i.e., they do not need to be monotonically ordered)::
 
     >>> y_pred = qrf.predict(X_test, quantiles=[0.5, 0.25, 0.75])
     >>> bool((y_pred[:, 0] >= y_pred[:, 1]).all())

quantile_forest/_quantile_forest.py

@@ -251,7 +251,7 @@ def _get_y_train_leaves_slice(
         Returns
         -------
         y_train_leaves_slice : array-like of shape \
-                (n_leaves, n_outputs, n_samples)
+                (n_leaves, n_outputs, n_indices)
             Mapping of training sample indices to tree's leaf nodes. Nodes with
             no samples (e.g., internal nodes) are empty. Internal nodes are
             included so that leaf node indices match their ``est.apply``
@@ -590,7 +590,7 @@ def _get_unsampled_indices(self, estimator, duplicates=None):
 
         Returns
         -------
-        unsampled_indices : array of shape (n_unsampled)
+        unsampled_indices : array of shape (n_unsampled,)
             Unsampled indices.
         """
         if not self.bootstrap:
@@ -686,7 +686,7 @@ def predict(
 
         Returns
         -------
-        y_pred : array of shape (n_samples, n_quantiles) or \
+        y_pred : array of shape (n_samples,) or (n_samples, n_quantiles) or \
                 (n_samples, n_outputs, n_quantiles)
             If quantiles is set to 'mean', then return ``E(Y | X)``. Else, for
             all quantiles, return ``y`` at ``q`` for which ``F(Y=y|x) = q``,
@@ -811,7 +811,7 @@ def quantile_ranks(
             ``dtype=np.float32``. If a sparse matrix is provided, it will be
             converted into a sparse ``csr_matrix``.
 
-        y : array-like of shape (n_samples) or (n_samples, n_outputs)
+        y : array-like of shape (n_samples,) or (n_samples, n_outputs)
             The target values for which to calculate ranks.
 
         kind : {"rank", "weak", "strict", "mean"}, default="rank"
@@ -848,7 +848,7 @@ def quantile_ranks(
 
         Returns
         -------
-        y_ranks : array of shape (n_samples) or (n_samples, n_outputs)
+        y_ranks : array of shape (n_samples,) or (n_samples, n_outputs)
             Quantile ranks in range [0, 1].
         """
         check_is_fitted(self)

quantile_forest/_quantile_forest_fast.pyx

@@ -126,7 +126,7 @@ cpdef vector[double] calc_quantile(
 
     Returns
     -------
-    out : array-like of shape (n_quantiles)
+    out : array-like of shape (n_quantiles,)
         Quantiles for `inputs` as floats. Empty list if empty inputs.
     """
     cdef double C = 1
@@ -255,7 +255,7 @@ cpdef vector[double] calc_weighted_quantile(
 
     Returns
     -------
-    out : array-like of shape (n_quantiles)
+    out : array-like of shape (n_quantiles,)
         Quantiles for `inputs` as floats. Empty list if empty inputs.
     """
     cdef double C = 1
@@ -572,7 +572,7 @@ cdef class QuantileForest:
         Parameters
         ----------
         quantiles : list of floats
-            Quantiles the in range [0, 1] to compute. -1 to compute mean.
+            Quantiles in the range [0, 1] to compute. -1 to compute mean.
 
         X_leaves : array-like of shape (n_samples, n_estimators)
             Target leaf node indices along samples and trees.

quantile_forest/_utils.pyx

@@ -67,7 +67,7 @@ cpdef vector[intp_t] generate_unsampled_indices(
 
     Parameters
     ----------
-    sample_indices : array-like of shape (n_samples)
+    sample_indices : array-like of shape (n_samples,)
         Sampled indices.
 
     n_total_samples : int
@@ -112,7 +112,7 @@ cpdef group_indices_by_value(cnp.ndarray[intp_t, ndim=1] a):
 
     Parameters
     ----------
-    a : array-like of shape (n_samples)
+    a : array-like of shape (n_samples,)
         Input array. The array is expected to contain integers, and the
         function will group the indices of elements with the same value.
 
@@ -198,7 +198,7 @@ cpdef map_indices_to_leaves(
 
     Parameters
     ----------
-    y_train_leaves_slice : array-like of shape (n_leaves, n_outputs, n_samples)
+    y_train_leaves_slice : array-like of shape (n_leaves, n_outputs, n_indices)
         Unpopulated mapping representing a list of nodes, each with a list of
         indices of the training samples residing at that node.
 
@@ -213,7 +213,7 @@ cpdef map_indices_to_leaves(
 
     Returns
     -------
-    y_train_leaves_slice : array-like of shape (n_leaves, n_outputs, n_samples)
+    y_train_leaves_slice : array-like of shape (n_leaves, n_outputs, n_indices)
         Populated mapping of training sample indices to leaf nodes. Nodes with
         no samples (e.g., internal nodes) are empty. Internal nodes are
         included so that leaf node indices match their ``est.apply`` outputs.