File extension of log file
ivangfalconi
Member Posts: 3 
Contributor I
<?xml version="1.0" encoding="UTF-8"?><process version="8.0.001">
<operator activated="true" class="write_csv" compatibility="8.0.001" expanded="true" height="82" name="Write CSV" width="90" x="45" y="289">
<parameter key="csv_file" value="problema"/>
<parameter key="column_separator" value=";"/>
<parameter key="write_attribute_names" value="true"/>
<parameter key="quote_nominal_values" value="true"/>
<parameter key="format_date_attributes" value="true"/>
<parameter key="append_to_file" value="false"/>
<parameter key="encoding" value="SYSTEM"/>
</operator>
</process>
Hi. Need help, please. Error message: "No log data table available". First, I named log file differently from the dataset (in "Create Log Data" operator) and creted a new folder for it. "Log" operator requests the path of folder. Did not work so I named log file with same name of dataset. Did not work either. What extension file should log file be? 
Best Answers
-
lionelderkrikor
RapidMiner Certified Analyst, Member Posts: 1,195 
Unicorn
Hi @ivangfalconi,
The XML process, you provided, is incomplete : It can be loaded.
Moreover we don't know what is your label attribute.
In view of the screenshots, you used the "predictive maintenance " process template.
Can you please share your whole process ?
Regards,
Lionel
1 -
ivangfalconi
Member Posts: 3 Contributor I
<?xml version="1.0" encoding="UTF-8"?><process version="8.0.001">
<operator activated="true" class="write_csv" compatibility="8.0.001" expanded="true" height="82" name="Write CSV" width="90" x="45" y="238">
<parameter key="column_separator" value=";"/>
<parameter key="write_attribute_names" value="true"/>
<parameter key="quote_nominal_values" value="true"/>
<parameter key="format_date_attributes" value="true"/>
<parameter key="append_to_file" value="false"/>
<parameter key="encoding" value="SYSTEM"/>
</operator>
</process>
<?xml version="1.0" encoding="UTF-8"?><process version="8.0.001">
<operator activated="true" class="write_csv" compatibility="8.0.001" expanded="true" height="82" name="Write CSV" width="90" x="380" y="442">
<parameter key="column_separator" value=","/>
<parameter key="write_attribute_names" value="true"/>
<parameter key="quote_nominal_values" value="true"/>
<parameter key="format_date_attributes" value="true"/>
<parameter key="append_to_file" value="false"/>
<parameter key="encoding" value="SYSTEM"/>
</operator>
</process>
<?xml version="1.0" encoding="UTF-8"?><process version="8.0.001">
<operator activated="true" class="write_csv" compatibility="8.0.001" expanded="true" height="82" name="Write CSV" width="90" x="380" y="391">
<parameter key="column_separator" value=","/>
<parameter key="write_attribute_names" value="true"/>
<parameter key="quote_nominal_values" value="true"/>
<parameter key="format_date_attributes" value="true"/>
<parameter key="append_to_file" value="false"/>
<parameter key="encoding" value="SYSTEM"/>
</operator>
</process>
<?xml version="1.0" encoding="UTF-8"?><process version="8.0.001">
<operator activated="true" class="write_csv" compatibility="8.0.001" expanded="true" height="82" name="Write CSV" width="90" x="380" y="391">
<parameter key="column_separator" value=","/>
<parameter key="write_attribute_names" value="true"/>
<parameter key="quote_nominal_values" value="true"/>
<parameter key="format_date_attributes" value="true"/>
<parameter key="append_to_file" value="false"/>
<parameter key="encoding" value="SYSTEM"/>
</operator>
</process>Thank you for your quick reply and apologize for my late reply. I am not sure if what I copied - complete process - is OK. It resembles the previous process already posted. In order to copy process for 1st part, I followed instructions from video. For remaining subprocesses I connected the write operator to the last operator. Is this correct?
And yes, ii is basically same model posted in the tab menu "Learn" except that I added a "Dicretize" operator in one of the Subprocesses.
1 -
lionelderkrikor
RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn
Hi @ivangfalconi,
It seems you are not using the good method to share your process :
You have just to follow these 2 steps :
1. Copy your XML code :
588582. Paste your XML code in the Insert Code pop up window then click "OK"
58553. That's it !
I hope it helps,
Regards,
Lionel
0
Answers
Hi again @ivangfalconi,
1. Log data table error.
I fix this error : You have to set the name of a Log operator. In your case, I choose the "Log model performance" Log operator which
is inside your Optimize Parameters operator.
2. Your process
After taking a look to your performance results, (accuracy = 0 and std deviation = 0 for all combinaisons of parameters), there are corrections to perform in your process. In a "classic" data-science project, you have to :
- Set a single attribute as label in your Set Role operator. (In your Set Role operator, all attributes are set as label).The label is the variable you want to predict.
- In a "classic" predictive maintenance project, the label is binary (failure / no failure). In your case, the V2 attribute can not a priori be a label without preprocessing, because it can take many values (integer type). It's an hypothesis, if this attribute is effectively the label , but you can apply it the Numerical to Binominal operator : below a threshold ==> failure / above a threshold => no failure (or vice-versa).
- All the other attributes have to be set as regular attribute in your Set Role operator. These attributes are the explicative variables. The model that you train establishes the relationships between these explicative variables and your label attribute.
- Delete the Set Role operator inside the Validation operator : You have to set the roles once.
Here the process :
NB : I disabled the "score branch" of your process because I have not your score dataset.
I hope it helps,
Regards,
Lionel