Skip to content

Commit

Permalink
Merge branch 'gh-pages' of github.com:mlr-org/mlr-tutorial into gh-pages
Browse files Browse the repository at this point in the history
  • Loading branch information
@schiffner
schiffner committed Nov 28, 2015
2 parents 95d2d34 + 9369246 commit 3f5e33f
Show file tree
Hide file tree
Showing 20 changed files with 271 additions and 269 deletions.
2 changes: 1 addition & 1 deletion devel/html/benchmark_experiments/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -439,7 +439,7 @@ <h2 id="example-one-task-two-learners-prediction-on-a-single-test-set">Example:
#&gt; Stratification: FALSE
#&gt; predict.type: response
#&gt; threshold:
#&gt; time (mean): 0.01
#&gt; time (mean): 0.02
#&gt; id truth response iter set
#&gt; 180 180 M M 1 test
#&gt; 100 100 M R 1 test
Expand Down
2 changes: 1 addition & 1 deletion devel/html/configureMlr/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -488,7 +488,7 @@ <h2 id="example-handling-errors-in-an-underlying-learning-method">Example: Handl
<pre><code class="r">## This call gives an error caused by the low number of observations in class &quot;virginica&quot;
train(&quot;classif.qda&quot;, task = iris.task, subset = 1:104)
#&gt; Error in qda.default(x, grouping, ...): some group is too small for 'qda'
#&gt; Timing stopped at: 0.004 0 0.005
#&gt; Timing stopped at: 0.005 0 0.005

## Turn learner errors into warnings
configureMlr(on.learner.error = &quot;warn&quot;)
Expand Down
26 changes: 14 additions & 12 deletions devel/html/cost_sensitive_classif/index.html
View file

Large diffs are not rendered by default.

4 changes: 2 additions & 2 deletions devel/html/create_learner/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -329,7 +329,7 @@

<li><a href="#regression">Regression</a></li>

<li><a href="#survival-analysis">Survival Analysis</a></li>
<li><a href="#survival-analysis">Survival analysis</a></li>

<li><a href="#clustering">Clustering</a></li>

Expand Down Expand Up @@ -516,7 +516,7 @@ <h2 id="regression">Regression</h2>

<p>Again most of the data is passed straight through to/from the train/predict
functions of the learner.</p>
<h2 id="survival-analysis">Survival Analysis</h2>
<h2 id="survival-analysis">Survival analysis</h2>
<p>For survival analysis, you have to return so-called linear predictors in order to compute
the default measure for this task type, the <a href="../measures/index.html">cindex</a> (for
<code>.learner$predict.type</code> == <code>"response"</code>). For <code>.learner$predict.type</code> == <code>"prob"</code>,
Expand Down
6 changes: 3 additions & 3 deletions devel/html/create_measure/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -369,15 +369,15 @@ <h2 id="performance-measures-and-aggregation-schemes">Performance measures and a
#&gt; {
#&gt; measureMSE(pred$data$truth, pred$data$response)
#&gt; }
#&gt; &lt;bytecode: 0x192b7000&gt;
#&gt; &lt;bytecode: 0x1f0f1820&gt;
#&gt; &lt;environment: namespace:mlr&gt;

measureMSE
#&gt; function (truth, response)
#&gt; {
#&gt; mean((response - truth)^2)
#&gt; }
#&gt; &lt;bytecode: 0xf8e6500&gt;
#&gt; &lt;bytecode: 0x111b2dd8&gt;
#&gt; &lt;environment: namespace:mlr&gt;
</code></pre>

Expand Down Expand Up @@ -412,7 +412,7 @@ <h2 id="performance-measures-and-aggregation-schemes">Performance measures and a
test.mean$fun
#&gt; function (task, perf.test, perf.train, measure, group, pred)
#&gt; mean(perf.test)
#&gt; &lt;bytecode: 0x1e7ad968&gt;
#&gt; &lt;bytecode: 0x1ab3c3d8&gt;
#&gt; &lt;environment: namespace:mlr&gt;
</code></pre>

Expand Down
26 changes: 13 additions & 13 deletions devel/html/feature_selection/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -504,13 +504,13 @@ <h3 id="tuning-the-size-of-the-feature-subset">Tuning the size of the feature su
<p>The performance of all percentage values visited during tuning is:</p>
<pre><code class="r">as.data.frame(res$opt.path)
#&gt; fw.perc mse.test.mean dob eol error.message exec.time
#&gt; 1 0.2 40.59578 1 NA &lt;NA&gt; 0.270
#&gt; 2 0.25 40.59578 2 NA &lt;NA&gt; 0.235
#&gt; 3 0.3 37.05592 3 NA &lt;NA&gt; 0.238
#&gt; 4 0.35 35.83712 4 NA &lt;NA&gt; 0.240
#&gt; 5 0.4 35.83712 5 NA &lt;NA&gt; 0.237
#&gt; 6 0.45 27.39955 6 NA &lt;NA&gt; 0.238
#&gt; 7 0.5 27.39955 7 NA &lt;NA&gt; 0.244
#&gt; 1 0.2 40.59578 1 NA &lt;NA&gt; 0.280
#&gt; 2 0.25 40.59578 2 NA &lt;NA&gt; 0.269
#&gt; 3 0.3 37.05592 3 NA &lt;NA&gt; 0.289
#&gt; 4 0.35 35.83712 4 NA &lt;NA&gt; 0.258
#&gt; 5 0.4 35.83712 5 NA &lt;NA&gt; 0.256
#&gt; 6 0.45 27.39955 6 NA &lt;NA&gt; 0.258
#&gt; 7 0.5 27.39955 7 NA &lt;NA&gt; 0.256
</code></pre>

<p>The optimal percentage and the corresponding performance can be accessed as follows:</p>
Expand Down Expand Up @@ -553,12 +553,12 @@ <h3 id="tuning-the-size-of-the-feature-subset">Tuning the size of the feature su
#&gt; Points on front: 13
head(as.data.frame(res$opt.path))
#&gt; fw.threshold fpr.test.mean fnr.test.mean dob eol error.message exec.time
#&gt; 1 0.4892321 0.3092818 0.2639033 1 NA &lt;NA&gt; 2.293
#&gt; 2 0.2481696 0.2045499 0.2319697 2 NA &lt;NA&gt; 2.296
#&gt; 3 0.7691875 0.5128000 0.3459740 3 NA &lt;NA&gt; 2.279
#&gt; 4 0.1470133 0.2045499 0.2319697 4 NA &lt;NA&gt; 2.348
#&gt; 5 0.5958241 0.5028216 0.5239538 5 NA &lt;NA&gt; 2.169
#&gt; 6 0.6892421 0.6323959 0.4480808 6 NA &lt;NA&gt; 2.091
#&gt; 1 0.4892321 0.3092818 0.2639033 1 NA &lt;NA&gt; 2.552
#&gt; 2 0.2481696 0.2045499 0.2319697 2 NA &lt;NA&gt; 2.617
#&gt; 3 0.7691875 0.5128000 0.3459740 3 NA &lt;NA&gt; 2.428
#&gt; 4 0.1470133 0.2045499 0.2319697 4 NA &lt;NA&gt; 2.610
#&gt; 5 0.5958241 0.5028216 0.5239538 5 NA &lt;NA&gt; 2.487
#&gt; 6 0.6892421 0.6323959 0.4480808 6 NA &lt;NA&gt; 2.469
</code></pre>

<p>The results can be visualized with function <a href="http://www.rdocumentation.org/packages/mlr/functions/plotTuneMultiCritResult.html">plotTuneMultiCritResult</a>.
Expand Down
2 changes: 1 addition & 1 deletion devel/html/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -419,5 +419,5 @@ <h4 class="modal-title" id="exampleModalLabel">Search</h4>

<!--
MkDocs version : 0.14.0
Build Date UTC : 2015-11-24 20:02:46.424248
Build Date UTC : 2015-11-27 16:39:29.130836
-->
58 changes: 29 additions & 29 deletions devel/html/integrated_learners/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -394,7 +394,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>size</code>has been set to 3 by default. Doing bagging training of <code>nnet</code>if set <code>bag=TRUE</code></td>
<td align="left"><code>size</code> has been set to 3 by default. Doing bagging training of <code>nnet</code> if set <code>bag=TRUE</code>.</td>
</tr>
<tr>
<td align="left"><strong>classif.bartMachine</strong> <br />bartmachine</td>
Expand Down Expand Up @@ -449,7 +449,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center">X</td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>xval</code>has been set to 0 by default for speed.</td>
<td align="left"><code>xval</code> has been set to 0 by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>classif.bst</strong> <br />bst</td>
Expand All @@ -460,7 +460,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">twoclass</td>
<td align="left">The argument <code>learner</code>has been renamed to <code>Learner</code>due to a name conflict with <code>setHyerPars</code> <code>Learner</code>has been set to <code>lm</code>by default.</td>
<td align="left">The argument <code>learner</code> has been renamed to <code>Learner</code> due to a name conflict with <code>setHyerPars</code>. <code>Learner</code> has been set to <code>lm</code> by default.</td>
</tr>
<tr>
<td align="left"><strong>classif.cforest</strong> <br />cforest</td>
Expand All @@ -482,7 +482,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">twoclass</td>
<td align="left"><code>centers</code>set to 2 by default</td>
<td align="left"><code>centers</code> set to 2 by default</td>
</tr>
<tr>
<td align="left"><strong>classif.ctree</strong> <br />ctree</td>
Expand All @@ -504,7 +504,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>output</code>set to <code>softmax</code>by default</td>
<td align="left"><code>output</code> set to <code>softmax</code> by default</td>
</tr>
<tr>
<td align="left"><strong>classif.dcSVM</strong> <br />dcSVM</td>
Expand Down Expand Up @@ -581,7 +581,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left">prob<br />twoclass</td>
<td align="left"><code>family</code>has been set to <code>Binomial()</code>by default. Maximum number of boosting iterations is set via 'mstop', the actual number used for prediction is controlled by 'm'.</td>
<td align="left"><code>family</code> has been set to <code>Binomial()</code> by default. Maximum number of boosting iterations is set via 'mstop', the actual number used for prediction is controlled by 'm'.</td>
</tr>
<tr>
<td align="left"><strong>classif.glmnet</strong> <br />glmnet</td>
Expand Down Expand Up @@ -669,7 +669,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">class.weights<br />multiclass<br />prob<br />twoclass</td>
<td align="left">Kernel parameters have to be passed directly and not by using the kpar list in ksvm. Note that <code>fit</code>has been set to <code>FALSE</code>by default for speed.</td>
<td align="left">Kernel parameters have to be passed directly and not by using the kpar list in ksvm. Note that <code>fit</code> has been set to <code>FALSE</code> by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>classif.lda</strong> <br />lda</td>
Expand Down Expand Up @@ -779,7 +779,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left">prob<br />twoclass</td>
<td align="left"><code>penalty</code>has been set to <em>lasso</em> and <code>lambda</code>to 0.1 by default.</td>
<td align="left"><code>penalty</code> has been set to <em>lasso</em> and <code>lambda</code> to 0.1 by default.</td>
</tr>
<tr>
<td align="left"><strong>classif.lssvm</strong> <br />lssvm</td>
Expand All @@ -790,7 +790,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />twoclass</td>
<td align="left"><code>fitted</code>has been set to <code>FALSE</code>by default for speed.</td>
<td align="left"><code>fitted</code> has been set to <code>FALSE</code> by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>classif.lvq1</strong> <br />lvq1</td>
Expand All @@ -812,7 +812,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>keep.fitted</code>has been set to <code>FALSE</code>by default for speed and we use start.method='lvq' for more robust behavior / less technical crashes</td>
<td align="left"><code>keep.fitted</code> has been set to <code>FALSE</code> by default for speed and we use start.method='lvq' for more robust behavior / less technical crashes</td>
</tr>
<tr>
<td align="left"><strong>classif.mlp</strong> <br />mlp</td>
Expand Down Expand Up @@ -856,7 +856,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">prob<br />twoclass</td>
<td align="left"><code>err.fct</code>has been set to <code>ce</code>to do classification.</td>
<td align="left"><code>err.fct</code> has been set to <code>ce</code> to do classification.</td>
</tr>
<tr>
<td align="left"><strong>classif.nnet</strong> <br />nnet</td>
Expand All @@ -867,7 +867,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>size</code>has been set to 3 by default.</td>
<td align="left"><code>size</code> has been set to 3 by default.</td>
</tr>
<tr>
<td align="left"><strong>classif.nnTrain</strong> <br />nn.train</td>
Expand All @@ -878,7 +878,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>output</code>set to <code>softmax</code>by default</td>
<td align="left"><code>output</code> set to <code>softmax</code> by default</td>
</tr>
<tr>
<td align="left"><strong>classif.nodeHarvest</strong> <br />nodeHarvest</td>
Expand Down Expand Up @@ -911,7 +911,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">prob<br />twoclass</td>
<td align="left">threshold for prediction (<code>threshold.predict</code> has been set to <code>1</code>by default</td>
<td align="left">threshold for prediction (<code>threshold.predict</code>) has been set to <code>1</code> by default</td>
</tr>
<tr>
<td align="left"><strong>classif.PART</strong> <br />part</td>
Expand Down Expand Up @@ -1010,7 +1010,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"></td>
<td align="left">By default, internal parallelization is switched off (num.threads = 1) and verbose output is disabled. Both settings are changeable.</td>
</tr>
<tr>
<td align="left"><strong>classif.rda</strong> <br />rda</td>
Expand All @@ -1021,7 +1021,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>estimate.error</code>has been set to <code>FALSE</code>by default for speed.</td>
<td align="left"><code>estimate.error</code> has been set to <code>FALSE</code> by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>classif.rFerns</strong> <br />rFerns</td>
Expand Down Expand Up @@ -1065,7 +1065,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center">X</td>
<td align="center">X</td>
<td align="left">multiclass<br />ordered<br />prob<br />twoclass</td>
<td align="left"><code>xval</code>has been set to 0 by default for speed.</td>
<td align="left"><code>xval</code> has been set to 0 by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>classif.rrlda</strong> <br />rrlda</td>
Expand All @@ -1087,7 +1087,7 @@ <h3 id="classification-70">Classification (70)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">multiclass<br />prob<br />twoclass</td>
<td align="left"><code>output</code>set to <code>softmax</code>by default</td>
<td align="left"><code>output</code> set to <code>softmax</code> by default</td>
</tr>
<tr>
<td align="left"><strong>classif.sda</strong> <br />sda</td>
Expand Down Expand Up @@ -1171,7 +1171,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left"></td>
<td align="left"><code>size</code>has been set to 3 by default.</td>
<td align="left"><code>size</code> has been set to 3 by default.</td>
</tr>
<tr>
<td align="left"><strong>regr.bartMachine</strong> <br />bartmachine</td>
Expand Down Expand Up @@ -1270,7 +1270,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left"></td>
<td align="left">The argument <code>learner</code>has been renamed to <code>Learner</code>due to a name conflict with <code>setHyerPars</code></td>
<td align="left">The argument <code>learner</code> has been renamed to <code>Learner</code> due to a name conflict with <code>setHyerPars</code></td>
</tr>
<tr>
<td align="left"><strong>regr.btgp</strong> <br />btgp</td>
Expand Down Expand Up @@ -1413,7 +1413,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center">X</td>
<td align="center">X</td>
<td align="left"></td>
<td align="left"><code>distribution</code>has been set to <em>gaussian</em> by default.</td>
<td align="left"><code>distribution</code> has been set to <em>gaussian</em> by default.</td>
</tr>
<tr>
<td align="left"><strong>regr.glmboost</strong> <br />glmboost</td>
Expand Down Expand Up @@ -1468,7 +1468,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left"></td>
<td align="left">Kernel parameters have to be passed directly and not by using the kpar list in ksvm. Note that <code>fit</code>has been set to <code>FALSE</code>by default for speed.</td>
<td align="left">Kernel parameters have to be passed directly and not by using the kpar list in ksvm. Note that <code>fit</code> has been set to <code>FALSE</code> by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>regr.LiblineaRL2L1SVR</strong> <br />liblinl2l1svr</td>
Expand Down Expand Up @@ -1534,7 +1534,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left"></td>
<td align="left"><code>size</code>has been set to 3 by default.</td>
<td align="left"><code>size</code> has been set to 3 by default.</td>
</tr>
<tr>
<td align="left"><strong>regr.nodeHarvest</strong> <br />nodeHarvest</td>
Expand Down Expand Up @@ -1622,7 +1622,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left"></td>
<td align="left"></td>
<td align="left">By default, internal parallelization is switched off (num.threads = 1) and verbose output is disabled. Both settings are changeable.</td>
</tr>
<tr>
<td align="left"><strong>regr.rknn</strong> <br />rknn</td>
Expand All @@ -1644,7 +1644,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center">X</td>
<td align="center">X</td>
<td align="left">ordered</td>
<td align="left"><code>xval</code>has been set to 0 by default for speed.</td>
<td align="left"><code>xval</code> has been set to 0 by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>regr.rsm</strong> <br />rsm</td>
Expand All @@ -1666,7 +1666,7 @@ <h3 id="regression-49">Regression (49)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left"></td>
<td align="left">Kernel parameters have to be passed directly and not by using the kpar list in rvm. Note that <code>fit</code>has been set to <code>FALSE</code>by default for speed.</td>
<td align="left">Kernel parameters have to be passed directly and not by using the kpar list in rvm. Note that <code>fit</code> has been set to <code>FALSE</code> by default for speed.</td>
</tr>
<tr>
<td align="left"><strong>regr.svm</strong> <br />svm</td>
Expand Down Expand Up @@ -1761,7 +1761,7 @@ <h3 id="survival-analysis-9">Survival analysis (9)</h3>
<td align="center"></td>
<td align="center">X</td>
<td align="left">ordered<br />rcens</td>
<td align="left"><code>family</code>has been set to <code>CoxPH()</code>by default. Maximum number of boosting iterations is set via 'mstop', the actual number used for prediction is controlled by 'm'.</td>
<td align="left"><code>family</code> has been set to <code>CoxPH()</code> by default. Maximum number of boosting iterations is set via 'mstop', the actual number used for prediction is controlled by 'm'.</td>
</tr>
<tr>
<td align="left"><strong>surv.glmnet</strong> <br />glmnet</td>
Expand Down Expand Up @@ -1805,7 +1805,7 @@ <h3 id="survival-analysis-9">Survival analysis (9)</h3>
<td align="center"></td>
<td align="center"></td>
<td align="left">prob<br />rcens</td>
<td align="left"></td>
<td align="left">By default, internal parallelization is switched off (num.threads = 1) and verbose output is disabled. Both settings are changeable.</td>
</tr>
<tr>
<td align="left"><strong>surv.rpart</strong> <br />rpart</td>
Expand All @@ -1816,7 +1816,7 @@ <h3 id="survival-analysis-9">Survival analysis (9)</h3>
<td align="center">X</td>
<td align="center">X</td>
<td align="left">ordered<br />rcens</td>
<td align="left"><code>xval</code>has been set to 0 by default for speed.</td>
<td align="left"><code>xval</code> has been set to 0 by default for speed.</td>
</tr>
</tbody>
</table>
Expand Down
Loading

0 comments on commit 3f5e33f

Please sign in to comment.