Update example plots

zillow · Sep 5, 2024 · 60d9b09 · 60d9b09
1 parent e6d7d63
commit 60d9b09
Show file tree

Hide file tree

Showing 13 changed files with 11 additions and 4 deletions.
diff --git a/docs/user_guide/fit_predict.rst b/docs/user_guide/fit_predict.rst
@@ -144,4 +144,4 @@ The predictions of a standard random forest can also be recovered from a quantil
 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`.
+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_custom_functions`.
diff --git a/examples/plot_quantile_conformalized.py → examples/plot_qrf_conformalized_intervals.py b/examples/plot_quantile_conformalized.py → examples/plot_qrf_conformalized_intervals.py
diff --git a/examples/plot_predict_custom.py → examples/plot_qrf_custom_functions.py b/examples/plot_predict_custom.py → examples/plot_qrf_custom_functions.py
diff --git a/examples/plot_quantile_extrapolation.py → examples/plot_qrf_extrapolation.py b/examples/plot_quantile_extrapolation.py → examples/plot_qrf_extrapolation.py
diff --git a/examples/plot_huggingface_model.py → examples/plot_qrf_huggingface_inference.py b/examples/plot_huggingface_model.py → examples/plot_qrf_huggingface_inference.py
diff --git a/examples/plot_quantile_interpolation.py → examples/plot_qrf_interpolation_methods.py b/examples/plot_quantile_interpolation.py → examples/plot_qrf_interpolation_methods.py
@@ -84,7 +84,7 @@ def plot_interpolation_predictions(df, legend):
     """Plot predictions by quantile interpolation methods."""
     # Slider for varying the prediction interval that determines the quantiles being interpolated.
     slider = alt.binding_range(name="Prediction Interval: ", min=0, max=1, step=0.01)
-    interval_val = alt.param(name="interval", value=0.9, bind=slider)
+    interval_val = alt.param(name="interval", value=0.8, bind=slider)
 
     click = alt.selection_point(bind="legend", fields=["method"], on="click")
 

diff --git a/examples/plot_quantile_multioutput.py → examples/plot_qrf_multitarget.py b/examples/plot_quantile_multioutput.py → examples/plot_qrf_multitarget.py
diff --git a/examples/plot_quantile_intervals.py → examples/plot_qrf_prediction_intervals.py b/examples/plot_quantile_intervals.py → examples/plot_qrf_prediction_intervals.py
diff --git a/examples/plot_quantile_example.py → examples/plot_qrf_predictions.py b/examples/plot_quantile_example.py → examples/plot_qrf_predictions.py
diff --git a/examples/plot_proximity_counts.py → examples/plot_qrf_proximity_counts.py b/examples/plot_proximity_counts.py → examples/plot_qrf_proximity_counts.py
@@ -24,7 +24,7 @@
 from quantile_forest import RandomForestQuantileRegressor
 
 random_state = np.random.RandomState(0)
-n_test_samples = 25
+n_test_samples = 50
 noise_std = 0.1
 
 pixel_dim = (8, 8)  # pixel dimensions (width and height)
@@ -76,7 +76,12 @@ def extract_floats(combined_df, scale=1000):
 # We set `max_samples_leaf=None` to ensure that every sample in the training
 # data is stored in the leaf nodes. By doing this, we allow the model to
 # consider all samples as potential candidates for proximity calculations.
-qrf = RandomForestQuantileRegressor(max_samples_leaf=None, random_state=random_state)
+qrf = RandomForestQuantileRegressor(
+    n_estimators=500,
+    max_features=1 / 3,
+    max_samples_leaf=None,
+    random_state=random_state,
+)
 
 # Fit the model to predict the non-noisy pixels from noisy pixels (i.e., to denoise).
 # We fit a single multi-target model, with each pixel treated as a distinct target.
@@ -126,6 +131,8 @@ def plot_digits_proximities(
     width=225,
 ):
     """Plot Digits dataset proximities for test samples."""
+    df = df[df["prox_idx"] < n_prox].copy()
+
     dim_x, dim_y = pixel_dim[0], pixel_dim[1]
     dgt_x, dgt_y = len(str(dim_x)), len(str(dim_y))
 

diff --git a/examples/plot_quantile_ranks.py → examples/plot_qrf_quantile_ranks.py b/examples/plot_quantile_ranks.py → examples/plot_qrf_quantile_ranks.py
diff --git a/examples/plot_treeshap_example.py → examples/plot_qrf_treeshap_explanations.py b/examples/plot_treeshap_example.py → examples/plot_qrf_treeshap_explanations.py
diff --git a/examples/plot_quantile_vs_standard.py → examples/plot_qrf_vs_rf_predictions.py b/examples/plot_quantile_vs_standard.py → examples/plot_qrf_vs_rf_predictions.py