using latest knowledge for hyperparameter search

src/single-core-regen/regen.py

@@ -30,10 +30,10 @@ data_settings = DataSettings(
 )
 
 pytorch_settings = PytorchSettings(
-    epochs=10,
+    epochs=10000,
     batchsize=2**10,
     device="cuda",
-    dataloader_workers=2,
+    dataloader_workers=12,
     dataloader_prefetch=4,
     summary_dir=".runs",
     write_every=2**5,
@@ -44,33 +44,31 @@ pytorch_settings = PytorchSettings(
 model_settings = ModelSettings(
     output_dim=2,
     # n_hidden_layers = (3, 8),
-    n_hidden_layers=(4, 6),  # study: single_core_regen_20241123_011232
-    n_hidden_nodes=(4,20),
-    # overrides={
-    #     "n_hidden_nodes_0": (14, 20),  # study: single_core_regen_20241123_011232
-    #     "n_hidden_nodes_1":  (8, 16),
-    #     "n_hidden_nodes_2":  (10, 16),
-    #     # "n_hidden_nodes_3":  (4, 20),  # study: single_core_regen_20241123_135749
-    #     "n_hidden_nodes_4": (2, 8),
-    #     "n_hidden_nodes_5": (10, 16),
-    #     },  
-    # model_activation_func    = ("ModReLU", "Mag", "Identity")
-    model_activation_func="Mag",  # study: single_core_regen_20241123_011232
+    n_hidden_layers=4,
+    overrides={
+        "n_hidden_nodes_0": 8,
+        "n_hidden_nodes_1": 6,
+        "n_hidden_nodes_2": 4,
+        "n_hidden_nodes_3": 8,
+    },
+    model_activation_func="Mag",
+    # satabsT0=(1e-6, 1),
 )
 
 optimizer_settings = OptimizerSettings(
     optimizer="Adam",
     # learning_rate       = (1e-5, 1e-1),
-    learning_rate=5e-4,
+    learning_rate=5e-3
+    # learning_rate=5e-4,
 )
 
 optuna_settings = OptunaSettings(
-    n_trials=512,
-    n_workers=14,
+    n_trials=1,
+    n_workers=1,
     timeout=3600,
-    directions=("maximize", "minimize"),
-    metrics_names=("neg_log_mse","n_nodes"),
-    limit_examples=True,
+    directions=("minimize",),
+    metrics_names=("mse",),
+    limit_examples=False,
     n_train_batches=500,
     # n_valid_batches = 100,
     storage="sqlite:///data/single_core_regen.db",