updated documentation

Description.props

@@ -26,7 +26,7 @@ Maintainer=Jan N. van Rijn <j.n.van.rijn@liacs.leidenuniv.nl>
 License=BSD-3-Clause
 
 # Description (required)
-Description=Allows sharing Weka experiments on OpenML. Stores for every experiment the classifier, the (serialized) models, the predictions and evaluation measures. Works well together in combination with the openml apiconnector (available on Maven Central, artifact id org.openml.apiconnector). If you found this package useful, please cite: J. N. van Rijn, Massively Collaborative Machine Learning, Leiden University, 2016. 
+Description=Allows sharing Weka experiments on OpenML. Stores for every experiment the classifier, the (serialized) models, the predictions and evaluation measures. Works well in combination with the openml apiconnector (available on Maven Central, artifact id org.openml.apiconnector). If you found this package useful, please cite: J. N. van Rijn, Massively Collaborative Machine Learning, Leiden University, 2016. 
 
 # Package URL for obtaining the package archive (required)
 PackageURL=https://github.com/openml/openml-weka/releases/download/openmlweka-0.9.5/OpenmlWeka.zip

README.md

@@ -2,16 +2,34 @@ Package for uploading Weka experiments to OpenML.
 
 Code example, download 100 (pre-selected) datasets and execute a REPTree on these.
 
-```
+```java
 public static void runTasksAndUpload() throws Exception {
-  OpenmlConnector openml = new OpenmlConnector("<FILL_IN_OPENML_API_KEY>");
-  Study s = openml.studyGet("OpenML100", "tasks"); 
-  Classifier tree = new REPTree(); 
-  for (Integer taskId : s.getTasks()) { 
-    Task t = openml.taskGet(taskId); 
-    Instances d = InstancesHelper.getDatasetFromTask(openml, t);
-    int runId = RunOpenmlJob.executeTask(openml, new WekaConfig(), taskId, tree);
+  // Fill in the API key (obtainable from your OpenML profile)
+  String apikey = "d488d8afd93b32331cf6ea9d7003d4c3";
+
+  // The WekaConfig module gives us the possibilities to enable or disable various Weka Specific options
+  WekaConfig config = new WekaConfig();
+
+  // Instantiate the OpenmlConnector object 
+  // requires artifact org.openml.apiconnector (version 1.0.14) from Maven central
+  OpenmlConnector openml = new OpenmlConnector(apikey);
+
+  // Download the OpenML object containing the `OpenML100' benchmark set
+  Study s = openml.studyGet("OpenML100", "tasks");
+
+  // Loop over all the tasks
+  for (Integer taskId : s.getTasks()) {
+    // create a Weka classifier to run on the task
+    Classifier tree = new NaiveBayes();
+
+    // execute the task (can take a while, depending on the classifier / dataset combination)
+    int runId = RunOpenmlJob.executeTask(openml, config, taskId, tree);
+
+    // After several minutes, the evaluation measures will be available on the server
+    System.out.println("Available on " + openml.getApiUrl() + "run/" + runId);
+
+    // Download the run from the server:
     Run run = openml.runGet(runId);
   }
-} 
+}
 ```

src/main/java/org/openml/weka/experiment/RunOpenmlJob.java

@@ -36,6 +36,7 @@ OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 
 import javax.swing.DefaultListModel;
 
+import org.openml.apiconnector.algorithms.Conversion;
 import org.openml.apiconnector.io.OpenmlConnector;
 import org.openml.apiconnector.xml.Task;
 import org.openml.weka.algorithm.WekaConfig;
@@ -87,7 +88,9 @@ public static int executeTask(OpenmlConnector openml, WekaConfig config, Integer
 
 		// set task
 		DefaultListModel<Task> tasks = new DefaultListModel<Task>();
-		tasks.add(0, openml.taskGet(task_id));
+		Task task = openml.taskGet(task_id);
+		Conversion.log("OK", "TaskGet", "Downloaded " + task.getTask_name());
+		tasks.add(0, task);
 		exp.setTasks(tasks);
 
 		// run the stuff