I got two question regarding the migration from Encog 3.0 to 3.1:
We got a lot of trained neuronal networks. These networks where trained with the Levenberg-Marquadt-Algorithm (LMA) and works where well.
The networks are stored in EQ files. We would like to upgrade to Encoq 3.1,
1. Can we use our trained networks further or have we to train them again?
2. Are serialized EQ files of the LMA in Encog 3.0 compatible with Encoq 3.1?