Update NAM-full configs (#463)

* Update NAM-full configs * Update NAM-full LSTM config
sdatkinson · Sep 17, 2024 · a24caf3 · a24caf3
1 parent 8fecd53
commit a24caf3
Show file tree

Hide file tree

Showing 3 changed files with 36 additions and 14 deletions.
diff --git a/nam_full_configs/data/single_pair.json b/nam_full_configs/data/single_pair.json
@@ -4,13 +4,13 @@
         "representative of this!"
     ],
     "train": {
-        "start": null,
-        "stop": -432000,
+        "start_seconds": null,
+        "stop_seconds": -9.0,
         "ny": 8192
     },
     "validation": {
-        "start": -432000,
-        "stop": null,
+        "start_seconds": -9.0,
+        "stop_seconds": null,
         "ny": null
     },
     "common": {

diff --git a/nam_full_configs/models/lstm.json b/nam_full_configs/models/lstm.json
@@ -18,19 +18,19 @@
         "name": "LSTM",
         "config": {
             "num_layers": 3,
-            "hidden_size": 24,
-            "train_burn_in": 4096,
-            "train_truncate": 512            
+            "hidden_size": 18,
+            "train_burn_in": 8192,
+            "train_truncate": null
         }
     },
     "loss": {
-        "val_loss": "mse",
-        "mask_first": 4096,    
-        "pre_emph_weight": 1.0,
-        "pre_emph_coef": 0.85
+        "val_loss": "esr",
+        "mask_first": 8192,
+        "pre_emph_mrstft_weight": 0.002,
+        "pre_emph_mrstft_coef": 0.85
     },
     "optimizer": {
-        "lr": 0.01
+        "lr": 0.008
     },
     "lr_scheduler": {
         "class": "ExponentialLR",

diff --git a/nam_full_configs/models/wavenet.json b/nam_full_configs/models/wavenet.json
@@ -13,7 +13,18 @@
                     "channels": 16,
                     "head_size": 8,
                     "kernel_size": 3,
-                    "dilations": [1,2,4,8,16,32,64,128,256,512],
+                    "dilations": [
+                        1,
+                        2,
+                        4,
+                        8,
+                        16,
+                        32,
+                        64,
+                        128,
+                        256,
+                        512
+                    ],
                     "activation": "Tanh",
                     "gated": false,
                     "head_bias": false
@@ -24,7 +35,18 @@
                     "channels": 8,
                     "head_size": 1,
                     "kernel_size": 3,
-                    "dilations": [1,2,4,8,16,32,64,128,256,512],
+                    "dilations": [
+                        1,
+                        2,
+                        4,
+                        8,
+                        16,
+                        32,
+                        64,
+                        128,
+                        256,
+                        512
+                    ],
                     "activation": "Tanh",
                     "gated": false,
                     "head_bias": true