[ci skip] Remove extra validation set (#80) ddc2f35

_downloads/a6916f06450964ef8d10eb5f311100d1/plot_01_survival_analysis.ipynb

@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "from sklearn.model_selection import train_test_split\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\nX_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2)"
+        "from sklearn.model_selection import train_test_split\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)"
       ]
     },
     {

_downloads/accdde75b290e8d01376db9bd16d592c/plot_01_survival_analysis.py

@@ -48,8 +48,7 @@
 # %%
 from sklearn.model_selection import train_test_split
 
-X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
-X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2)
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)
 
 # %%
 #

_sources/auto_examples/plot_01_survival_analysis.rst.txt

@@ -187,14 +187,13 @@ In this dataset, approximately 42% of the data is censored..
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 49-54
+.. GENERATED FROM PYTHON SOURCE LINES 49-53
 
 .. code-block:: Python
 
     from sklearn.model_selection import train_test_split
 
-    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
-    X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2)
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)
 
 
 
@@ -203,7 +202,7 @@ In this dataset, approximately 42% of the data is censored..
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 55-71
+.. GENERATED FROM PYTHON SOURCE LINES 54-70
 
 Using SurvivalBoost to estimate the survival function
 -----------------------------------------------------
@@ -222,7 +221,7 @@ SurvivalBoost is a scikit-learn compatible model which expects a covariates data
 (or array-like) ``X``, and a target dataframe ``y`` with columns "event" and
 "duration". This allows SurvivalBoost to estimate the survival function :math:`S`.
 
-.. GENERATED FROM PYTHON SOURCE LINES 72-78
+.. GENERATED FROM PYTHON SOURCE LINES 71-77
 
 .. code-block:: Python
 
@@ -649,15 +648,15 @@ SurvivalBoost is a scikit-learn compatible model which expects a covariates data
     <br />
     <br />
 
-.. GENERATED FROM PYTHON SOURCE LINES 79-84
+.. GENERATED FROM PYTHON SOURCE LINES 78-83
 
 SurvivalBoost can then predict the survival function for each patient,
 according to some time grid of horizons.
 **The time grid is learned during fit but can be passed during prediction**
 with the parameter ``times``.
 When ``times`` is set to ``None``, the model will used the learned time grid.
 
-.. GENERATED FROM PYTHON SOURCE LINES 85-94
+.. GENERATED FROM PYTHON SOURCE LINES 84-93
 
 .. code-block:: Python
 
@@ -677,11 +676,11 @@ When ``times`` is set to ``None``, the model will used the learned time grid.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 95-96
+.. GENERATED FROM PYTHON SOURCE LINES 94-95
 
 Let's plot the estimated survival function for some patients.
 
-.. GENERATED FROM PYTHON SOURCE LINES 96-127
+.. GENERATED FROM PYTHON SOURCE LINES 95-126
 
 .. code-block:: Python
 
@@ -728,7 +727,7 @@ Let's plot the estimated survival function for some patients.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 128-138
+.. GENERATED FROM PYTHON SOURCE LINES 127-137
 
 Measuring features impact on predictions
 ----------------------------------------
@@ -741,7 +740,7 @@ features to eliminate correlations.
 We create a synthetic dataset where age (``x8``) is resampled to reduce
 confounder bias.
 
-.. GENERATED FROM PYTHON SOURCE LINES 139-187
+.. GENERATED FROM PYTHON SOURCE LINES 138-186
 
 .. code-block:: Python
 
@@ -805,15 +804,15 @@ confounder bias.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 188-193
+.. GENERATED FROM PYTHON SOURCE LINES 187-192
 
 Unsurprisingly, the cumulative incidence of death mostly increases with age.
 We can do the same thing with chemotherapy treatement.
 
 Let's create a synthetic dataset where chemotherapy (``x6``)
 alternates between 0 and 1.
 
-.. GENERATED FROM PYTHON SOURCE LINES 194-235
+.. GENERATED FROM PYTHON SOURCE LINES 193-234
 
 .. code-block:: Python
 
@@ -870,7 +869,7 @@ alternates between 0 and 1.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 236-304
+.. GENERATED FROM PYTHON SOURCE LINES 235-303
 
 People treated with chemotherapy likely have more advanced stages of cancer, which is
 reflected by the lower estimated survival function. This serves as a reminder that
@@ -941,7 +940,7 @@ summarize the Brier score in time:
     \mathrm{BS(t)} dt
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 305-315
+.. GENERATED FROM PYTHON SOURCE LINES 304-314
 
 .. code-block:: Python
 
@@ -963,17 +962,17 @@ summarize the Brier score in time:
 
  .. code-block:: none
 
-    IBS for SurvivalBoost: 0.1382
+    IBS for SurvivalBoost: 0.1439
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 316-318
+.. GENERATED FROM PYTHON SOURCE LINES 315-317
 
 We can compare this to the Integrated Brier score of a simple Kaplan-Meier estimator,
 which doesn't take the patient features into account.
 
-.. GENERATED FROM PYTHON SOURCE LINES 319-339
+.. GENERATED FROM PYTHON SOURCE LINES 318-338
 
 .. code-block:: Python
 
@@ -1005,16 +1004,16 @@ which doesn't take the patient features into account.
 
  .. code-block:: none
 
-    IBS for Kaplan-Meier: 0.1566
+    IBS for Kaplan-Meier: 0.1653
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 340-341
+.. GENERATED FROM PYTHON SOURCE LINES 339-340
 
 Let's also compute the concordance index for both the Kaplan-Meier and SurvivalBoost.
 
-.. GENERATED FROM PYTHON SOURCE LINES 344-353
+.. GENERATED FROM PYTHON SOURCE LINES 343-352
 
 .. code-block:: Python
 
@@ -1040,13 +1039,13 @@ Let's also compute the concordance index for both the Kaplan-Meier and SurvivalB
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 354-357
+.. GENERATED FROM PYTHON SOURCE LINES 353-356
 
 0.5 corresponds to random chance, which makes sense as the Kaplan-Meier estimator
 doesn't depend on the patient features.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 358-365
+.. GENERATED FROM PYTHON SOURCE LINES 357-364
 
 .. code-block:: Python
 
@@ -1073,7 +1072,7 @@ doesn't depend on the patient features.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 6.993 seconds)
+   **Total running time of the script:** (0 minutes 7.376 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_01_survival_analysis.py:

_sources/auto_examples/plot_02_marginal_cumulative_incidence_estimation.rst.txt

@@ -277,15 +277,15 @@ theoretical CIFs:
 
  .. code-block:: none
 
-    Integrated theoretical any event survival curve in 0.662 s
-    SurvivalBoost fit: 2.690 s
-    SurvivalBoost prediction: 2.927 s
-    Integrated theoretical cumulative incidence curve for event 1 in 2.988 s
-    Aalen-Johansen for event 1 fit in 4.937 s
-    Integrated theoretical cumulative incidence curve for event 2 in 5.032 s
-    Aalen-Johansen for event 2 fit in 5.018 s
-    Integrated theoretical cumulative incidence curve for event 3 in 5.096 s
-    Aalen-Johansen for event 3 fit in 4.976 s
+    Integrated theoretical any event survival curve in 0.614 s
+    SurvivalBoost fit: 2.766 s
+    SurvivalBoost prediction: 2.911 s
+    Integrated theoretical cumulative incidence curve for event 1 in 2.971 s
+    Aalen-Johansen for event 1 fit in 5.112 s
+    Integrated theoretical cumulative incidence curve for event 2 in 5.210 s
+    Aalen-Johansen for event 2 fit in 5.024 s
+    Integrated theoretical cumulative incidence curve for event 3 in 5.102 s
+    Aalen-Johansen for event 3 fit in 4.967 s
 
 
 
@@ -328,15 +328,15 @@ of censoring.
 
  .. code-block:: none
 
-    Integrated theoretical any event survival curve in 0.591 s
-    SurvivalBoost fit: 2.705 s
-    SurvivalBoost prediction: 2.940 s
-    Integrated theoretical cumulative incidence curve for event 1 in 3.000 s
-    Aalen-Johansen for event 1 fit in 4.967 s
-    Integrated theoretical cumulative incidence curve for event 2 in 5.058 s
-    Aalen-Johansen for event 2 fit in 4.967 s
-    Integrated theoretical cumulative incidence curve for event 3 in 5.045 s
-    Aalen-Johansen for event 3 fit in 5.035 s
+    Integrated theoretical any event survival curve in 0.576 s
+    SurvivalBoost fit: 2.708 s
+    SurvivalBoost prediction: 2.914 s
+    Integrated theoretical cumulative incidence curve for event 1 in 2.974 s
+    Aalen-Johansen for event 1 fit in 4.936 s
+    Integrated theoretical cumulative incidence curve for event 2 in 5.027 s
+    Aalen-Johansen for event 2 fit in 4.917 s
+    Integrated theoretical cumulative incidence curve for event 3 in 4.995 s
+    Aalen-Johansen for event 3 fit in 4.988 s
 
 
 
@@ -360,7 +360,7 @@ the large time horizons:
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 43.187 seconds)
+   **Total running time of the script:** (0 minutes 43.202 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_02_marginal_cumulative_incidence_estimation.py:

auto_examples/plot_01_survival_analysis.html

@@ -506,8 +506,7 @@
 </div>
 <div class="highlight-Python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">sklearn.model_selection</span> <span class="kn">import</span> <a href="https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split" class="sphx-glr-backref-module-sklearn-model_selection sphx-glr-backref-type-py-function"><span class="n">train_test_split</span></a>
 
-<span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <a href="https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split" class="sphx-glr-backref-module-sklearn-model_selection sphx-glr-backref-type-py-function"><span class="n">train_test_split</span></a><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">test_size</span><span class="o">=</span><span class="mf">0.2</span><span class="p">)</span>
-<span class="n">X_train</span><span class="p">,</span> <span class="n">X_val</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_val</span> <span class="o">=</span> <a href="https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split" class="sphx-glr-backref-module-sklearn-model_selection sphx-glr-backref-type-py-function"><span class="n">train_test_split</span></a><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">test_size</span><span class="o">=</span><span class="mf">0.2</span><span class="p">)</span>
+<span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <a href="https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split" class="sphx-glr-backref-module-sklearn-model_selection sphx-glr-backref-type-py-function"><span class="n">train_test_split</span></a><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">test_size</span><span class="o">=</span><span class="mf">0.3</span><span class="p">)</span>
 </pre></div>
 </div>
 <section id="using-survivalboost-to-estimate-the-survival-function">
@@ -1157,7 +1156,7 @@ <h2>Survival model evaluation<a class="headerlink" href="#survival-model-evaluat
 <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;IBS for SurvivalBoost: </span><span class="si">{</span><span class="n">ibs_survboost</span><span class="si">:</span><span class="s2">.4f</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>IBS for SurvivalBoost: 0.1382
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>IBS for SurvivalBoost: 0.1439
 </pre></div>
 </div>
 <p>We can compare this to the Integrated Brier score of a simple Kaplan-Meier estimator,
@@ -1182,7 +1181,7 @@ <h2>Survival model evaluation<a class="headerlink" href="#survival-model-evaluat
 <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;IBS for Kaplan-Meier: </span><span class="si">{</span><span class="n">ibs_km</span><span class="si">:</span><span class="s2">.4f</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>IBS for Kaplan-Meier: 0.1566
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>IBS for Kaplan-Meier: 0.1653
 </pre></div>
 </div>
 <p>Let’s also compute the concordance index for both the Kaplan-Meier and SurvivalBoost.</p>
@@ -1212,7 +1211,7 @@ <h2>Survival model evaluation<a class="headerlink" href="#survival-model-evaluat
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Concordance index for SurvivalBoost: 0.67
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 6.993 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 7.376 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-plot-01-survival-analysis-py">
 <div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
 <p><a class="reference download internal" download="" href="../_downloads/a6916f06450964ef8d10eb5f311100d1/plot_01_survival_analysis.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_01_survival_analysis.ipynb</span></code></a></p>

auto_examples/plot_02_marginal_cumulative_incidence_estimation.html

@@ -559,15 +559,15 @@ <h2>CIFs estimated on uncensored data<a class="headerlink" href="#cifs-estimated
 <span class="p">)</span>
 </pre></div>
 </div>
-<img src="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_001.png" srcset="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_001.png" alt="Cause-specific cumulative incidence functions (0.0% censoring), Event 1, Event 2, Event 3" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Integrated theoretical any event survival curve in 0.662 s
-SurvivalBoost fit: 2.690 s
-SurvivalBoost prediction: 2.927 s
-Integrated theoretical cumulative incidence curve for event 1 in 2.988 s
-Aalen-Johansen for event 1 fit in 4.937 s
-Integrated theoretical cumulative incidence curve for event 2 in 5.032 s
-Aalen-Johansen for event 2 fit in 5.018 s
-Integrated theoretical cumulative incidence curve for event 3 in 5.096 s
-Aalen-Johansen for event 3 fit in 4.976 s
+<img src="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_001.png" srcset="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_001.png" alt="Cause-specific cumulative incidence functions (0.0% censoring), Event 1, Event 2, Event 3" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Integrated theoretical any event survival curve in 0.614 s
+SurvivalBoost fit: 2.766 s
+SurvivalBoost prediction: 2.911 s
+Integrated theoretical cumulative incidence curve for event 1 in 2.971 s
+Aalen-Johansen for event 1 fit in 5.112 s
+Integrated theoretical cumulative incidence curve for event 2 in 5.210 s
+Aalen-Johansen for event 2 fit in 5.024 s
+Integrated theoretical cumulative incidence curve for event 3 in 5.102 s
+Aalen-Johansen for event 3 fit in 4.967 s
 </pre></div>
 </div>
 </section>
@@ -590,15 +590,15 @@ <h2>CIFs estimated on censored data<a class="headerlink" href="#cifs-estimated-o
 <span class="n">plot_cumulative_incidence_functions</span><span class="p">(</span><a href="../generated/hazardous.SurvivalBoost.html#hazardous.SurvivalBoost" title="hazardous.SurvivalBoost" class="sphx-glr-backref-module-hazardous sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">survival_boost</span></a><span class="o">=</span><a href="../generated/hazardous.SurvivalBoost.html#hazardous.SurvivalBoost" title="hazardous.SurvivalBoost" class="sphx-glr-backref-module-hazardous sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">survival_boost</span></a><span class="p">,</span> <span class="n">aj</span><span class="o">=</span><span class="n">aj</span><span class="p">,</span> <span class="n">y</span><span class="o">=</span><span class="n">y_censored</span><span class="p">)</span>
 </pre></div>
 </div>
-<img src="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_002.png" srcset="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_002.png" alt="Cause-specific cumulative incidence functions (40.4% censoring), Event 1, Event 2, Event 3" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Integrated theoretical any event survival curve in 0.591 s
-SurvivalBoost fit: 2.705 s
-SurvivalBoost prediction: 2.940 s
-Integrated theoretical cumulative incidence curve for event 1 in 3.000 s
-Aalen-Johansen for event 1 fit in 4.967 s
-Integrated theoretical cumulative incidence curve for event 2 in 5.058 s
-Aalen-Johansen for event 2 fit in 4.967 s
-Integrated theoretical cumulative incidence curve for event 3 in 5.045 s
-Aalen-Johansen for event 3 fit in 5.035 s
+<img src="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_002.png" srcset="../_images/sphx_glr_plot_02_marginal_cumulative_incidence_estimation_002.png" alt="Cause-specific cumulative incidence functions (40.4% censoring), Event 1, Event 2, Event 3" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Integrated theoretical any event survival curve in 0.576 s
+SurvivalBoost fit: 2.708 s
+SurvivalBoost prediction: 2.914 s
+Integrated theoretical cumulative incidence curve for event 1 in 2.974 s
+Aalen-Johansen for event 1 fit in 4.936 s
+Integrated theoretical cumulative incidence curve for event 2 in 5.027 s
+Aalen-Johansen for event 2 fit in 4.917 s
+Integrated theoretical cumulative incidence curve for event 3 in 4.995 s
+Aalen-Johansen for event 3 fit in 4.988 s
 </pre></div>
 </div>
 <p>Note that the Aalen-Johansen estimator is unbiased and empirically recovers
@@ -613,7 +613,7 @@ <h2>CIFs estimated on censored data<a class="headerlink" href="#cifs-estimated-o
 <p>Alternatively, we could try to enable a monotonicity constraint at training
 time, however, in practice this often causes a sever over-estimation bias for
 the large time horizons:</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 43.187 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 43.202 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-plot-02-marginal-cumulative-incidence-estimation-py">
 <div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
 <p><a class="reference download internal" download="" href="../_downloads/8da6be5df74b4f584c69dbcd5de4f948/plot_02_marginal_cumulative_incidence_estimation.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_02_marginal_cumulative_incidence_estimation.ipynb</span></code></a></p>