Working example for time series estimation

carl/learning/calibration.py

@@ -95,8 +95,8 @@ def fit(self, X, y, sample_weight=None):
         * `self` [object]:
             `self`.
         """
-        # Check inputs
-        X, y = check_X_y(X, y)
+        # Check inputs. Avoid this in order to use RNNs
+        # X, y = check_X_y(X, y)
 
         # Convert y
         label_encoder = LabelEncoder()

carl/ratios/utils.py

@@ -0,0 +1,73 @@
+"""This module implements some utility functions."""
+
+# Carl is free software; you can redistribute it and/or modify it
+# under the terms of the Revised BSD License; see LICENSE file for
+# more details.
+
+import numpy as np
+from .classifier import ClassifierRatio
+import matplotlib.pyplot as plt
+
+
+def plot_score_(ratio, axis, reals, labels):
+    cal_num, cal_den = (ratio.classifier_.calibrators_[0].calibrator0,
+                        ratio.classifier_.calibrators_[0].calibrator1)
+    axis.plot(reals, cal_num.pdf(reals.reshape(-1, 1)),
+              label="p(s_num), num~{0}".format(labels[0]))
+    axis.plot(reals, cal_den.pdf(reals.reshape(-1, 1)),
+              label="p(s_den), den~{0}".format(labels[1]))
+    axis.legend(frameon=False)
+
+
+def plot_scores(classifier_ratios, num_den_labels=None, save_file=None):
+    """
+    Plot score plots for a list of classifier ratios,
+    useful to check training and calibration quality
+
+    Parameters
+    ----------
+    * `classifier_ratios` [list of `ClassifierRatio`]:
+        List of ClassifierRatio to plot.
+
+    * `num_den_labels` [list of (num_label, den_label) tuples]:
+        List of numeratod and denominator labels for each
+        ratio in `classifier_ratios`
+
+    """
+
+    if len(classifier_ratios) == 0:
+        raise ValueError
+    for classifier in classifier_ratios:
+        if not isinstance(classifier, ClassifierRatio):
+            raise ValueError
+
+    num_ratios = len(classifier_ratios)
+    if num_den_labels is None:
+        num_den_labels = [(l, 0) for l in
+                          range(len(classifier_ratios))]
+    else:
+        # Ensure format for numbers
+        def rounds(n): return map(lambda x: round(x, 1), n)
+        num_den_labels = [(rounds(x[0]), rounds(x[1])) for x in num_den_labels]
+
+    reals = np.linspace(0, 1)
+
+    if num_ratios <= 3:
+        f, axarr = plt.subplots(1, num_ratios, sharex=True, sharey=True,
+                                figsize=(12, 5))
+        for k, (ratio, pos) in enumerate(zip(classifier_ratios,
+                                             num_den_labels)):
+            plot_score_(ratio, axarr[k], reals, labels=pos)
+    else:
+        f, axarr = plt.subplots((num_ratios // 3) + (num_ratios % 3 != 0), 3,
+                                sharex=True, sharey=True)
+        for k, (ratio, pos) in enumerate(zip(classifier_ratios, 
+                                             num_den_labels)):
+            plot_score_(ratio, axarr[k // 3, k % 3], reals, labels=pos)
+    f.subplots_adjust(hspace=0.05, wspace=0.05)
+    if save_file is None:
+        plt.show()
+    else:
+        plt.savefig(save_file)
+        plt.close()
+        plt.clf()