Hello,
in which version do you get the error? Is that printed out by 1.5.3 or
by 1.3?
Did you get an exception? If so please post the full stack trace.
In the course of the development the training data used to train the
English models
was extended and improved when errors came up.
To give it a fair comparison you should use the same training data for
both versions.
Jörn
On 02/25/2014 12:40 AM, Barry Olddog wrote:
That was the error message from opennlp/model/GenericModelReader.checkModelType
signature
I'm using the 1.5.3 binary from the mirror at LSU, and I'm using the
downloaded models from sourceforge to test how the tools perform
differently from version 1.3.
Many thanks if anyone can give me a clue about how to fix.
Barry
