What would cause a Network Hash & Encoder Hash mismatch from training to inference, when I haven’t actually changed anything in the blueprints?
Everything was running fine until this morning; all of a sudden regardless how many times i repackage and retrain, there’s always a mismatch between the network and the schema.
`[2025.11.20-12.05.32:355][890]LogLearning: Error: Policy_0: Encoder Network Asset provided during Setup is incompatible with Schema. Network hash is -2085981630 vs Schema hash -403027718.`
Nothing was changed at all during any of these times, so i find it quite odd that it’s started happening today.
EDIT/UPDATE: Solved, Network asset file must’ve corrupted somehow during an editor crash. Created a new asset and it’s fixed and running again.