Cobblemon Hostile Neural Networks Compat

This is a data pack for Minecraft 1.21.1 that adds over one thousand compat files so that Hostile Neural Networks accepts Cobblemon entities as valid data models. It has only been tested on Neoforge but should work on other mod loaders. In order to use this, you should install it to the datapacks folder, as well as to the world folder; if you are using something like Paxi or Global Packs (recommended,) you should place it there too in order to make sure the datapack is applied server-wide. This seems to be the common methodology of installation for HNN compats. It is required to be installed on both the Client and the Server.

Each data model simulation cost is tuned based on a cobblemon's MST and weight as well as its rarity and catch difficulty. The simulation drops also additionally add a small bit of XP candy depending on the rarity and type gems (1-2) per simulation based on the cobblemon's primary and secondary type. The simulated drops are defaults based on Cobblemon's existing loot table.

Most Cobblemon base item are overworld_prediction, however, Cobblemon with the "legendary" label are the end_prediction. Because Cobblemon are not like standard mobs that spawn in a predictible manner (they are often hard to find, and some packs make legendaries unique server-wide,) one mob kill instantly upgrades the data model to the Advanced tier.

To balance this, Superior and Self-Aware data model tiers are adjusted to require more cycles (2k, and 3k respectively.) Unfortunately there was no way to make the model start out at Basic tier without requiring at least one kill due to format restrictions.

This is intended for use with Cobblemon 1.8.1. It also supports Pickmin and Kenji's Yoshi custom cobblemon.

You can also use a mod like Poke Power with this mod, though it might supply TOO much power. In that case, you can download the version marked PP_1, which raises the simulation cost by a factor of ~183x.