openfront-rl / ppo_v83 /manifest.json
djmango's picture
Sync manifest.json (99b21d39cf34)
76e7e29 verified
Raw
History Blame
1.39 kB
{
"format": "oftrain-safetensors",
"manifest_schema_version": 1,
"architecture": {
"name": "oftrain-policy",
"schema_version": 2,
"dimensions": {
"grid_channels": 89,
"fine_grid_channels": 90,
"player_features": 21,
"scalars": 11,
"local_planes": 5,
"actions": 21,
"build_types": 7,
"nuke_types": 5,
"quantity_params": 2,
"grid_tower_channels": 256,
"grid_tower_blocks": 4,
"hidden": 512,
"player_hidden": 128,
"local_hidden": 64,
"transformer_layers": 2,
"attention_heads": 4
},
"recurrent": {
"cell": "gru",
"hidden_size": 256,
"context_schema": "action-outcome-v1",
"context_features": 14,
"context_embedding": 128,
"bptt_length": 16,
"rollout_length": 32,
"residual_initialization": "zero-output-projection",
"hidden_reset_policy": "episode_done"
}
},
"autoencoders": {
"fine": {
"ref": "../weights/ae/ae_v31_d8c32.encoder.safetensors",
"format": "safetensors"
},
"coarse": {
"ref": "../weights/ae/ae_v31_d16c32.encoder.safetensors",
"format": "safetensors"
}
},
"checkpoint": {
"weights": "latest.safetensors",
"state": "latest.state.json"
},
"update": 356,
"stage": 7,
"curriculum_schedule": "v8.3",
"reward_profile": "v8.3-closeout-v1"
}