Frequently asked questions

• Using EOP as a library and let my code use it

• Adding a new language model to MaltParser

• 401 Unauthorized error using using myProject's code examples

• ERROR EOPRunner: Could not perform training

---

Using EOP as a library and let my code use it

A number of EOP functionalities are accessible both via the EOP Application Program Interface (API) and Command Line Interface (CLI). The Step by Step Tutorial teaches you how to do that by concrete examples.

Adding a new language model to MaltParser

Dependency trees in the EOP are produced by MaltPaser that is available for the 3 languages: English, German and Italian. In case you need to port MaltParser into a new language you have to produce a new language model. The following page explains how it can be done: [Add new language models to MaltParser](Add new language models to MaltParser)

401 Unauthorized using myProject's code examples

myProject-EOPv{version}.tar.gz contains some [examples of code](Hello World! code and additional examples) that you can use to become more confident with the EOP. It could happen that when you import it into Eclipse, Eclipse produces a 401 Unauthorized error. Generally it is due to one of the following:

• TreeTagger has not been installed
• The TreeTagger artifacts version reported in the pom.xml file of myProject is different from that installed with TreeTagger and that is in the user local maven repository (e.g., /home/user_name/.m2/).

The documentation [here](Step-by-Step, TreeTagger-Installation) reports the correct procedure to install TreeTagger and how let it work with the EOP.

ERROR EOPRunner: Could not perform training

EOPRunner is the Java class to be used when you want to use the EOP from the command line. When you call EOPRunner so that it pre-processes the data set and then train/test an EDA, EOPRunner saves the pre-processed files into a temporary directory (it is set in the configuration file of the used EDA); it could happen that if you do not remember to remove the pre-processed files of a previuous run, EOPRunner will use both the previous and current files for training/testing. As a consequence in the best case you will have unexpected results while in other cases, when for example those files have been annotated with a different pre-processing, EOPRunner could stop with an error: Could not perform training.