Update docs

zillow · Apr 10, 2024 · 9100b59 · 9100b59
1 parent 2132641
commit 9100b59
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 8 deletions.
diff --git a/docs/user_guide/fit_predict.rst b/docs/user_guide/fit_predict.rst
@@ -141,7 +141,7 @@ The predictions of a standard random forest can also be recovered from a quantil
     >>> np.allclose(y_pred_rf, y_pred_qrf)
     True
 
-Predicting with User-Specified Functions
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+User-Specified Functions
+~~~~~~~~~~~~~~~~~~~~~~~~
 
 While a QRF is designed to estimate quantiles from the empirical distribution calculated for each sample, in many cases it may be useful to use the empirical distribution to calculate other quantities of interest. For more details, see :ref:`gallery_plot_predict_custom`.
diff --git a/examples/plot_predict_custom.py b/examples/plot_predict_custom.py
@@ -1,6 +1,6 @@
 """
-Calculating User-Specified Functions with QRFs
-==============================================
+Computing User-Specified Functions with QRFs
+============================================
 
 An example that demonstrates a way of extracting the empirical distribution
 from a quantile regression forest (QRF) for one or more samples in order to
@@ -46,7 +46,7 @@ def predict(reg, X, quantiles=0.5, what=None):
     if what is None:
         return reg.predict(X, quantiles=quantiles)
 
-    # Get the complete set of proximities for each sample.
+    # Get the complete set of proximities (training indices) for each sample.
     proximities = reg.proximity_counts(X)
 
     # Retrieve the unsorted training responses from the model (stored in sorted order).
@@ -95,7 +95,7 @@ def plot_ecdf(df):
 
     circles = (
         alt.Chart(df)
-        .mark_circle(color="#006aff")
+        .mark_circle(color="#006aff", opacity=1, size=50)
         .encode(
             x=alt.X("y_value", title="Response Value"),
             y=alt.Y("probability", title="Probability"),
@@ -105,7 +105,7 @@ def plot_ecdf(df):
 
     lines = (
         alt.Chart(df)
-        .mark_line(color="#006aff")
+        .mark_line(color="#006aff", size=2)
         .encode(
             x=alt.X("y_value", title="Response Value"),
             x2=alt.X2("y_value2"),
@@ -114,7 +114,11 @@ def plot_ecdf(df):
         )
     )
 
-    chart = (circles + lines).properties(height=400, width=650)
+    chart = (circles + lines).properties(
+        height=400,
+        width=650,
+        title="Empirical Cumulative Distribution Function (ECDF) Plot",
+    )
     return chart