Bug message when running Keras: Extraction of nominal example value for non-nominal attribute 'RUL'

pxkst970

Hello, I have the following bug below when I tried to run Keras (Deep Neural Net 3 Dense layer with 1 Dropout layer). I've also attached the screenshot of my process. Note that I tried the same process with the different datasets but got no issue. Both datasets predict the numerical value (the label is number, not category). If anyone knows it's a bug or I did something wrong ... please ... please help.

Thank you!

====================================================================================

例外:com.rapidminer.example。AttributeTypeException

Message: Extraction of nominal example value for non-nominal attribute 'RUL' is not possible.

Stack trace:

com.rapidminer.example.Example.getNominalValue(Example.java:98)

com.rapidminer.operator.performance.SimpleCriterion.countExample(SimpleCriterion.java:93)

com.rapidminer.operator.performance.AbstractPerformanceEvaluator.evaluate(AbstractPerformanceEvaluator.java:470)

com.rapidminer.operator.performance.AbstractPerformanceEvaluator.evaluate(AbstractPerformanceEvaluator.java:393)

com.rapidminer.operator.performance.AbstractPerformanceEvaluator.doWork(AbstractPerformanceEvaluator.java:256)

com.rapidminer.operator.Operator.execute(Operator.java:1032)

com.rapidminer.operator.execution.SimpleUnitExecutor.execute(SimpleUnitExecutor.java:77)

com.rapidminer.operator.ExecutionUnit$2.run(ExecutionUnit.java:812)

com.rapidminer.operator.ExecutionUnit$2.run(ExecutionUnit.java:807)

java.security.AccessController.doPrivileged(Native Method)

com.rapidminer.operator.ExecutionUnit.execute(ExecutionUnit.java:807)

com.rapidminer.operator.OperatorChain.doWork(OperatorChain.java:423)

com.rapidminer.operator.Operator.execute(Operator.java:1032)

com.rapidminer.Process.executeRoot(Process.java:1378)

com.rapidminer.Process.lambda$executeRootInPool$5(Process.java:1357)

com.rapidminer.studio.concurrency.internal.AbstractConcurrencyContext$AdaptedCallable.exec(AbstractConcurrencyContext.java:328)

java.util.concurrent.ForkJoinTask.doExec(ForkJoinTask.java:289)

java.util.concurrent.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1056)

java.util.concurrent.ForkJoinPool.runWorker(ForkJoinPool.java:1692)

java.util.concurrent.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:157)

rfuentealba

Hello pxkst970,

Do you mind to share with us the actual XML of your process and the code that is failing? I'm pretty sure it's an issue with data cleaning, but better safe than sorry. If you can't publish it, you may use the private messaging for us to check.

All the best,

Rodrigo.

pxkst970

Hellorfuentealba,
Sure, no problem. I can share the actual XML with you below.
I suspected that for some reason, the Keras layer transformed my label (or prediction attribute) into nominal data. This is the regression problem and my label is numerical data. There is certainly no problem with data cleaning because I use this same dataset to do one layer ANN (Artificial Neural Net) in RapidMiner before I tried Keras. The problem occurred when applying the model from Keras into the test dataset.

I've also attached the layer in the Keras process for you as well. The problem might come from the last layer. As I understand the last layer should act as an output layer and should be a dense layer with no activation function, right? Please help to correct me if I'm wrong.

=========================================================================

<参数可y="logverbosity" value="init"/>

<参数可y="random_seed" value="2001"/>

<参数可y="send_mail" value="never"/>

<参数可y="notification_email" value=""/>

<参数可y="process_duration_for_mail" value="30"/>

<参数可y="encoding" value="SYSTEM"/>

<参数可y="csv_file" value="C:\Users\User\Desktop\NDSU\Research\6. Data\Aero_Engine\Experiment_FD002\1_Results_Final\Selected_Input_Final\Aero_Engine_1_FD002_z-normailized_Ori.csv"/>

<参数可y="column_separators" value=","/>

<参数可y="trim_lines" value="false"/>

<参数可y="use_quotes" value="true"/>

<参数可y="quotes_character" value="""/>

<参数可y="escape_character" value="\"/>

<参数可y="skip_comments" value="true"/>

<参数可y="comment_characters" value="//www.turtlecreekpls.com/community/discussion/56824/#"/>

<参数可y="starting_row" value="1"/>

<参数可y="parse_numbers" value="true"/>

<参数可y="decimal_character" value="."/>

<参数可y="grouped_digits" value="false"/>

<参数可y="grouping_character" value=","/>

<参数可y="infinity_representation" value=""/>

<参数可y="date_format" value=""/>

<参数可y="first_row_as_names" value="true"/>

<参数可y="0" value="Name"/>

<参数可y="time_zone" value="SYSTEM"/>

<参数可y="locale" value="English (United States)"/>

<参数可y="encoding" value="UTF-8"/>

<参数可y="read_all_values_as_polynominal" value="false"/>

<参数可y="0" value="T2.true.real.attribute"/>

<参数可y="1" value="T24.true.real.attribute"/>

<参数可y="2" value="T30.true.real.attribute"/>

<参数可y="3" value="T50.true.real.attribute"/>

<参数可y="4" value="P2.true.real.attribute"/>

<参数可y="5" value="P15.true.real.attribute"/>

<参数可y="6" value="P30.true.real.attribute"/>

<参数可y="7" value="Nf.true.real.attribute"/>

<参数可y="8" value="Nc.true.real.attribute"/>

<参数可y="9" value="epr.true.real.attribute"/>

<参数可y="10" value="Ps30.true.real.attribute"/>

<参数可y="11" value="phi.true.real.attribute"/>

<参数可y="12" value="NRF.true.real.attribute"/>

<参数可y="13" value="NRc.true.real.attribute"/>

<参数可y="14" value="BPR.true.real.attribute"/>

<参数可y="15" value="farB.true.real.attribute"/>

<参数可y="16" value="htBleed.true.real.attribute"/>

<参数可y="17" value="Nf_dmd.true.real.attribute"/>

<参数可y="18" value="PCNfR_dmd.true.real.attribute"/>

<参数可y="19" value="W31.true.real.attribute"/>

<参数可y="20" value="W32.true.real.attribute"/>

<参数可y="21" value="time_cycle.true.integer.id"/>

<参数可y="22" value="RUL.true.real.label"/>

<参数可y="read_not_matching_values_as_missings" value="false"/>

<参数可y="datamanagement" value="double_array"/>

<参数可y="data_management" value="auto"/>

<参数可y="input shape" value="(21,)"/>

<参数可y="loss" value="mean_squared_error"/>

<参数可y="optimizer" value="Adam"/>

<参数可y="learning rate" value="0.001"/>

<参数可y="momentum" value="0.0"/>

<参数可y="rho" value="0.9"/>

<参数可y="beta 1" value="0.999"/>

<参数可y="beta 2" value="0.999"/>

<参数可y="epsilon" value="1.0E-8"/>

<参数可y="decay" value="0.0"/>

<参数可y="schedule decay" value="0.004"/>

<参数可y="Nesterov" value="false"/>

<参数可y="use metric" value="false"/>

<参数可y="epochs" value="526"/>

<参数可y="batch size" value="1"/>

<参数可y="callbacks" value="TensorBoard(log_dir='./logs', histogram_freq=0, write_graph=True, write_images=False, embeddings_freq=0, embeddings_layer_names=None, embeddings_metadata=None)"/>

<参数键=“冗长”价值= " 1 " / >

<参数可y="validation split" value="0.0"/>

<参数可y="shuffle" value="false"/>

<参数可y="fix seed" value="false"/>

<参数可y="random seed" value="0"/>

<参数可y="layer_type" value="Dense"/>

<参数可y="no_units" value="12"/>

<参数可y="activation_function" value="'relu'"/>

<参数可y="use_bias" value="true"/>

<参数可y="kernel_initializer" value="VarianceScaling(scale=1.0, mode='fan_in', distribution='normal', seed=None)"/>

<参数可y="bias_initializer" value="Zeros()"/>

<参数可y="kernel_regularizer" value="None"/>

<参数可y="bias_regularizer" value="None"/>

<参数可y="activity_regularizer" value="None"/>

<参数可y="kernel_constraint" value="None"/>

<参数可y="bias_constraint" value="None"/>

<参数键=“率”价值= " 0.1 " / >

<参数可y="noise_shape" value="None"/>

<参数可y="seed" value="None"/>

<参数可y="target_shape" value="(1, 1)"/>

<参数可y="dims" value="1.1"/>

<参数可y="repetition_factor" value="2"/>

<参数可y="function" value="None"/>

<参数可y="l1" value="0.0"/>

<参数可y="l2" value="0.0"/>

<参数可y="mask_value" value="0.0"/>

<参数可y="layer_type" value="Dense"/>

<参数可y="no_units" value="12"/>

<参数可y="activation_function" value="'relu'"/>

<参数可y="use_bias" value="true"/>

<参数可y="kernel_initializer" value="VarianceScaling(scale=1.0, mode='fan_in', distribution='normal', seed=None)"/>

<参数可y="bias_initializer" value="Zeros()"/>

<参数可y="kernel_regularizer" value="None"/>

<参数可y="bias_regularizer" value="None"/>

<参数可y="activity_regularizer" value="None"/>

<参数可y="kernel_constraint" value="None"/>

<参数可y="bias_constraint" value="None"/>

<参数键=“率”价值= " 0.1 " / >

<参数可y="noise_shape" value="None"/>

<参数可y="seed" value="None"/>

<参数可y="target_shape" value="(1, 1)"/>

<参数可y="dims" value="1.1"/>

<参数可y="repetition_factor" value="2"/>

<参数可y="function" value="None"/>

<参数可y="l1" value="0.0"/>

<参数可y="l2" value="0.0"/>

<参数可y="mask_value" value="0.0"/>

<参数可y="layer_type" value="Dense"/>

<参数可y="no_units" value="12"/>

<参数可y="activation_function" value="'relu'"/>

<参数可y="use_bias" value="true"/>

<参数可y="kernel_initializer" value="VarianceScaling(scale=1.0, mode='fan_in', distribution='normal', seed=None)"/>

<参数可y="bias_initializer" value="Zeros()"/>

<参数可y="kernel_regularizer" value="None"/>

<参数可y="bias_regularizer" value="None"/>

<参数可y="activity_regularizer" value="None"/>

<参数可y="kernel_constraint" value="None"/>

<参数可y="bias_constraint" value="None"/>

<参数键=“率”价值= " 0.1 " / >

<参数可y="noise_shape" value="None"/>

<参数可y="seed" value="None"/>

<参数可y="target_shape" value="(1, 1)"/>

<参数可y="dims" value="1.1"/>

<参数可y="repetition_factor" value="2"/>

<参数可y="function" value="None"/>

<参数可y="l1" value="0.0"/>

<参数可y="l2" value="0.0"/>

<参数可y="mask_value" value="0.0"/>

<参数可y="layer_type" value="Dropout"/>

<参数可y="no_units" value="1"/>

<参数可y="activation_function" value="None"/>

<参数可y="use_bias" value="true"/>

<参数可y="kernel_initializer" value="glorot_uniform(seed=None)"/>

<参数可y="bias_initializer" value="Zeros()"/>

<参数可y="kernel_regularizer" value="None"/>

<参数可y="bias_regularizer" value="None"/>

<参数可y="activity_regularizer" value="None"/>

<参数可y="kernel_constraint" value="None"/>

<参数可y="bias_constraint" value="None"/>

<参数可y="rate" value="0.25"/>

<参数可y="noise_shape" value="None"/>

<参数可y="seed" value="None"/>

<参数可y="target_shape" value=""/>

<参数可y="dims" value="1.1"/>

<参数可y="repetition_factor" value="1"/>

<参数可y="function" value="None"/>

<参数可y="l1" value="0.0"/>

<参数可y="l2" value="0.0"/>

<参数可y="mask_value" value="0.0"/>

<参数可y="layer_type" value="Dense"/>

<参数可y="no_units" value="1"/>

<参数可y="activation_function" value="None"/>

<参数可y="use_bias" value="true"/>

<参数可y="kernel_initializer" value="VarianceScaling(scale=1.0, mode='fan_in', distribution='normal', seed=None)"/>

<参数可y="bias_initializer" value="Zeros()"/>

<参数可y="kernel_regularizer" value="None"/>

<参数可y="bias_regularizer" value="None"/>

<参数可y="activity_regularizer" value="None"/>

<参数可y="kernel_constraint" value="None"/>

<参数可y="bias_constraint" value="None"/>

<参数键=“率”价值= " 0.1 " / >

<参数可y="noise_shape" value="None"/>

<参数可y="seed" value="None"/>

<参数可y="target_shape" value="1.1"/>

<参数可y="dims" value="1.1"/>

<参数可y="repetition_factor" value="2"/>

<参数可y="function" value="None"/>

<参数可y="l1" value="0.0"/>

<参数可y="l2" value="0.0"/>

<参数可y="mask_value" value="0.0"/>

<参数可y="excel_file" value="C:\Users\User\Desktop\NDSU\Research\6. Data\Aero_Engine\Experiment_FD002\1_Results_Final_Latest\Aero_Keras_DNN_Loss_Function.xlsx"/>

<参数可y="file_format" value="xlsx"/>

<参数可y="sheet_name" value="RapidMiner Data"/>

<参数可y="date_format" value="yyyy-MM-dd HH:mm:ss"/>

<参数可y="number_format" value="#.0"/>

<参数可y="encoding" value="SYSTEM"/>

<参数可y="csv_file" value="C:\Users\User\Desktop\NDSU\Research\6. Data\Aero_Engine\Experiment_FD002\1_Results_Final\Selected_Input_Final\Test_Data_Engine-46_to_60_z-normalized_Ori.csv"/>

<参数可y="column_separators" value=","/>

<参数可y="trim_lines" value="false"/>

<参数可y="use_quotes" value="true"/>

<参数可y="quotes_character" value="""/>

<参数可y="escape_character" value="\"/>

<参数可y="skip_comments" value="true"/>

<参数可y="comment_characters" value="//www.turtlecreekpls.com/community/discussion/56824/#"/>

<参数可y="starting_row" value="1"/>

<参数可y="parse_numbers" value="true"/>

<参数可y="decimal_character" value="."/>

<参数可y="grouped_digits" value="false"/>

<参数可y="grouping_character" value=","/>

<参数可y="infinity_representation" value=""/>

<参数可y="date_format" value=""/>

<参数可y="first_row_as_names" value="true"/>

<参数可y="0" value="Name"/>

<参数可y="time_zone" value="SYSTEM"/>

<参数可y="locale" value="English (United States)"/>

<参数可y="encoding" value="UTF-8"/>

<参数可y="read_all_values_as_polynominal" value="false"/>

<参数可y="0" value="T2.true.real.attribute"/>

<参数可y="1" value="T24.true.real.attribute"/>

<参数可y="2" value="T30.true.real.attribute"/>

<参数可y="3" value="T50.true.real.attribute"/>

<参数可y="4" value="P2.true.real.attribute"/>

<参数可y="5" value="P15.true.real.attribute"/>

<参数可y="6" value="P30.true.real.attribute"/>

<参数可y="7" value="Nf.true.real.attribute"/>

<参数可y="8" value="Nc.true.real.attribute"/>

<参数可y="9" value="epr.true.real.attribute"/>

<参数可y="10" value="Ps30.true.real.attribute"/>

<参数可y="11" value="phi.true.real.attribute"/>

<参数可y="12" value="NRF.true.real.attribute"/>

<参数可y="13" value="NRc.true.real.attribute"/>

<参数可y="14" value="BPR.true.real.attribute"/>

<参数可y="15" value="farB.true.real.attribute"/>

<参数可y="16" value="htBleed.true.real.attribute"/>

<参数可y="17" value="Nf_dmd.true.real.attribute"/>

<参数可y="18" value="PCNfR_dmd.true.real.attribute"/>

<参数可y="19" value="W31.true.real.attribute"/>

<参数可y="20" value="W32.true.real.attribute"/>

<参数可y="21" value="time_cycle.true.integer.id"/>

<参数可y="22" value="RUL.true.real.label"/>

<参数可y="read_not_matching_values_as_missings" value="false"/>

<参数可y="datamanagement" value="double_array"/>

<参数可y="data_management" value="auto"/>

<参数可y="batch_size" value="1"/>

<参数可y="verbose" value="0"/>

<参数可y="main_criterion" value="root_mean_squared_error"/>

<参数可y="root_mean_squared_error" value="true"/>

<参数可y="absolute_error" value="false"/>

<参数可y="relative_error" value="false"/>

<参数可y="relative_error_lenient" value="false"/>

<参数可y="relative_error_strict" value="false"/>

<参数可y="normalized_absolute_error" value="false"/>

<参数可y="root_relative_squared_error" value="false"/>

<参数可y="squared_error" value="false"/>

<参数可y="correlation" value="false"/>

<参数可y="squared_correlation" value="false"/>

<参数可y="prediction_average" value="false"/>

<参数可y="spearman_rho" value="false"/>

<参数可y="kendall_tau" value="false"/>

<参数可y="skip_undefined_labels" value="true"/>

<参数可y="use_example_weights" value="true"/>

<参数可y="excel_file" value="C:\Users\User\Desktop\NDSU\Research\6. Data\Aero_Engine\Experiment_FD002\1_Results_Final_Latest\Aero_Keras_DNN_Result.xlsx"/>

<参数可y="file_format" value="xlsx"/>

<参数可y="sheet_name" value="RapidMiner Data"/>

<参数可y="date_format" value="yyyy-MM-dd HH:mm:ss"/>

<参数可y="number_format" value="#.0"/>

<参数可y="encoding" value="SYSTEM"/>

< portSpacing端口= " sink_result 3”间隔= " 0 " / >