Upload master_config.json
Browse files- master_config.json +40 -0
master_config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"seed": 42,
|
| 3 |
+
"output_dir": "./mindgames_optimized",
|
| 4 |
+
"F": 3,
|
| 5 |
+
"U": 20,
|
| 6 |
+
"R": 5,
|
| 7 |
+
"hidden": 192,
|
| 8 |
+
"gnn_layers": 3,
|
| 9 |
+
"gnn_heads": 6,
|
| 10 |
+
"n_strat": 6,
|
| 11 |
+
"ppo": {
|
| 12 |
+
"episodes": 1000,
|
| 13 |
+
"update_steps": 512,
|
| 14 |
+
"batch_size": 128,
|
| 15 |
+
"epochs": 4,
|
| 16 |
+
"clip": 0.2,
|
| 17 |
+
"vf_coef": 0.5,
|
| 18 |
+
"ent_coef": 0.01,
|
| 19 |
+
"gamma": 0.99,
|
| 20 |
+
"gae_lambda": 0.95,
|
| 21 |
+
"lr_slow": 3e-05,
|
| 22 |
+
"lr_fast": 0.0006,
|
| 23 |
+
"opponent_hist_len": 16,
|
| 24 |
+
"log_every": 50,
|
| 25 |
+
"eval_every": 250,
|
| 26 |
+
"eval_games": 200,
|
| 27 |
+
"meta_every": 4,
|
| 28 |
+
"meta_lr": 0.001
|
| 29 |
+
},
|
| 30 |
+
"pref_states": 800,
|
| 31 |
+
"pref_rollouts_per_pair": 4,
|
| 32 |
+
"do_sft": 1,
|
| 33 |
+
"do_dpo": 1,
|
| 34 |
+
"sft_steps": 150,
|
| 35 |
+
"dpo_steps": 200,
|
| 36 |
+
"distill_dataset_size": 2000,
|
| 37 |
+
"distill_steps": 400,
|
| 38 |
+
"distill_batch_size": 64,
|
| 39 |
+
"eval_games": 200
|
| 40 |
+
}
|