Pulling the pytorch stuff out of simple linear models as didn't add a…

…nything and made things complex

cca_zoo/_base.py

@@ -153,9 +153,12 @@ def pairwise_correlations(self, views: Iterable[np.ndarray], **kwargs):
                     ]
                 )
             )
-        all_corrs = np.array(all_corrs).reshape(
-            (self.n_views_, self.n_views_, self.latent_dimensions)
-        )
+        try:
+            all_corrs = np.array(all_corrs).reshape(
+                (self.n_views_, self.n_views_, self.latent_dimensions)
+            )
+        except:
+            print()
         return all_corrs
 
     def score(self, views: Iterable[np.ndarray], y=None, **kwargs):

docs/source/examples/plot_validation.py

@@ -6,69 +6,21 @@
 permutation testing, learning curves, and cross-validation.
 """
 
-# %%
 # Import libraries
-import matplotlib.pyplot as plt
 import numpy as np
+import matplotlib.pyplot as plt
 from sklearn.model_selection import ShuffleSplit, KFold
 
 from cca_zoo.data.simulated import LinearSimulatedData
 from cca_zoo.linear import CCA
 from cca_zoo.model_selection import learning_curve, permutation_test_score
 
-# %%
-# Data
-# ------
-# We set the random seed for reproducibility
-np.random.seed(42)
-
-# We generate a linear dataset with 200 samples, 15 features per view,
-# 3 latent dimensions and different correlations between the views
-n = 200
-p = 15
-q = 15
-latent_dims = 3
-correlations = [0.9, 0.5, 0.1]
-
-(X, Y) = LinearSimulatedData(
-    view_features=[p, q], latent_dims=latent_dims, correlation=correlations
-).sample(n)
-
-# %%
-# Permutation Testing
-# -------------------
-# Permutation testing is a way to assess the significance of the model performance by comparing it with the performance on permuted data.
-# We use a CCA model with 3 latent dimensions and a 2-fold cross-validation scheme.
-model = CCA(latent_dimensions=latent_dims)
-cv = KFold(2, shuffle=True, random_state=0)
-
-# We use permutation_test_score to compute the score on the original data and on 100 permutations of the data.
-score, perm_scores, pvalue = permutation_test_score(
-    model, (X, Y), cv=cv, n_permutations=100
-)
-
-# %%
-# We plot the histogram of the permuted scores and the score on the original data for each dimension.
-fig, ax = plt.subplots(latent_dims, figsize=[12, 8])
-for k in range(latent_dims):
-    ax[k].hist(perm_scores[k])
-    ax[k].axvline(score[k], ls="--", color="r")
-    score_label = f"Score on original\ndata: {score[k]:.2f}\n(p-value: {pvalue[k]:.3f})"
-    ax[k].text(0.05, 0.8, score_label, fontsize=12, transform=ax[k].transAxes)
-    ax[k].set_xlabel("Correlation")
-    _ = ax[k].set_ylabel("Frequency")
-    ax[k].set_title(f"Dimension {k + 1}")
-plt.tight_layout()
-plt.show()
-
-
-# %%
-# Learning Curves
-# -------------------
-
-import matplotlib.pyplot as plt
-import numpy as np
-
+np.random.seed(42)  # We set the random seed for reproducibility
+n = 250  # number of samples
+p = 15  # features in view 1
+q = 15  # features in view 2
+latent_dims = 1  # latent dimensions
+correlations = [0.9]  # correlations between views
 
 def plot_learning_curve(
     estimator,
@@ -219,15 +171,39 @@ def plot_learning_curve(
 
     return plt
 
+# Data generation
+(X, Y) = LinearSimulatedData(
+    view_features=[p, q], latent_dims=latent_dims, correlation=correlations
+).sample(n)
 
+# Permutation Testing
+model = CCA(latent_dimensions=latent_dims)
+cv = KFold(2, shuffle=True, random_state=0)
+score, perm_scores, pvalue = permutation_test_score(
+    model, (X, Y), cv=cv, n_permutations=100
+)
+
+# Permutation Test Visualization
+fig, ax = plt.subplots(latent_dims, figsize=[12, 8])
+for k in range(latent_dims):
+    ax.hist(perm_scores)
+    ax.axvline(score, ls="--", color="r")
+    score_label = f"Score on original\ndata: {score:.2f}\n(p-value: {pvalue:.3f})"
+    ax.text(0.05, 0.8, score_label, fontsize=12, transform=ax.transAxes)
+    ax.set_xlabel("Correlation")
+    _ = ax.set_ylabel("Frequency")
+    ax.set_title(f"Dimension {k + 1}")
+plt.tight_layout()
+plt.show()
+
+
+# Learning Curves
 fig, axes = plt.subplots(3, 1, figsize=(10, 15))
 
 title = "Learning Curves CCA"
-# Cross validation with 50 iterations to get smoother mean test and train
-# score curves, each time with 20% data randomly selected as a validation set.
 cv = ShuffleSplit(n_splits=50, test_size=0.2, random_state=0)
-
 model = CCA()
+
 plot_learning_curve(model, title, (X, Y), axes=axes, ylim=(0.7, 1.01), cv=cv, n_jobs=4)
 
 plt.show()