Check layer types for Optimizer construction (#10598)

* Check layer types for Optimizer construction

* Duplicate class

src/transformers/trainer.py

@@ -80,6 +80,7 @@ from .trainer_pt_utils import (
     SequentialDistributedSampler,
     distributed_broadcast_scalars,
     distributed_concat,
+    get_parameter_names,
     nested_concat,
     nested_detach,
     nested_numpify,
@@ -613,14 +614,15 @@ class Trainer:
         Trainer's init through :obj:`optimizers`, or subclass and override this method in a subclass.
         """
         if self.optimizer is None:
-            no_decay = ["bias", "LayerNorm.weight"]
+            decay_parameters = get_parameter_names(self.model, [torch.nn.LayerNorm])
+            decay_parameters = [name for name in decay_parameters if "bias" not in name]
             optimizer_grouped_parameters = [
                 {
-                    "params": [p for n, p in self.model.named_parameters() if not any(nd in n for nd in no_decay)],
+                    "params": [p for n, p in self.model.named_parameters() if n in decay_parameters],
                     "weight_decay": self.args.weight_decay,
                 },
                 {
-                    "params": [p for n, p in self.model.named_parameters() if any(nd in n for nd in no_decay)],
+                    "params": [p for n, p in self.model.named_parameters() if n not in decay_parameters],
                     "weight_decay": 0.0,
                 },
             ]

src/transformers/trainer_pt_utils.py

@@ -672,3 +672,19 @@ def save_state(self):
 
     path = os.path.join(self.args.output_dir, "trainer_state.json")
     self.state.save_to_json(path)
+
+
+def get_parameter_names(model, forbidden_layer_types):
+    """
+    Returns the names of the model parameters that are not inside a forbidden layer.
+    """
+    result = []
+    for name, child in model.named_children():
+        result += [
+            f"{name}.{n}"
+            for n in get_parameter_names(child, forbidden_layer_types)
+            if not isinstance(child, tuple(forbidden_layer_types))
+        ]
+    # Add model specific parameters (defined with nn.Parameter) since they are not in any child.
+    result += list(model._parameters.keys())
+    return result

tests/test_trainer.py

@@ -193,6 +193,20 @@ if is_torch_available():
             loss = torch.nn.functional.mse_loss(y, labels)
             return (loss, y, y) if self.double_output else (loss, y)
 
+    class TstLayer(torch.nn.Module):
+        def __init__(self, hidden_size):
+            super().__init__()
+            self.linear1 = torch.nn.Linear(hidden_size, hidden_size)
+            self.ln1 = torch.nn.LayerNorm(hidden_size)
+            self.linear2 = torch.nn.Linear(hidden_size, hidden_size)
+            self.ln2 = torch.nn.LayerNorm(hidden_size)
+            self.bias = torch.nn.Parameter(torch.zeros(hidden_size))
+
+        def forward(self, x):
+            h = self.ln1(torch.nn.functional.relu(self.linear1(x)))
+            h = torch.nn.functional.relu(self.linear2(x))
+            return self.ln2(x + h + self.bias)
+
     def get_regression_trainer(a=0, b=0, double_output=False, train_len=64, eval_len=64, pretrained=True, **kwargs):
         label_names = kwargs.get("label_names", None)
         train_dataset = RegressionDataset(length=train_len, label_names=label_names)
@@ -991,6 +1005,18 @@ class TrainerIntegrationTest(unittest.TestCase):
         # perfect world: fp32_init/2 == fp16_eval
         self.assertAlmostEqual(fp16_eval, fp32_init / 2, delta=5_000)
 
+    def test_no_wd_param_group(self):
+        model = torch.nn.Sequential(TstLayer(128), torch.nn.ModuleList([TstLayer(128), TstLayer(128)]))
+        trainer = Trainer(model=model)
+        trainer.create_optimizer_and_scheduler(10)
+        # fmt: off
+        wd_names = ['0.linear1.weight', '0.linear2.weight', '1.0.linear1.weight', '1.0.linear2.weight', '1.1.linear1.weight', '1.1.linear2.weight']
+        # fmt: on
+        wd_params = [p for n, p in model.named_parameters() if n in wd_names]
+        no_wd_params = [p for n, p in model.named_parameters() if n not in wd_names]
+        self.assertListEqual(trainer.optimizer.param_groups[0]["params"], wd_params)
+        self.assertListEqual(trainer.optimizer.param_groups[1]["params"], no_wd_params)
+
 
 @require_torch
 @require_optuna

tests/test_trainer_utils.py

@@ -30,8 +30,23 @@ if is_torch_available():
         DistributedTensorGatherer,
         LabelSmoother,
         LengthGroupedSampler,
+        get_parameter_names,
     )
 
+    class TstLayer(torch.nn.Module):
+        def __init__(self, hidden_size):
+            super().__init__()
+            self.linear1 = torch.nn.Linear(hidden_size, hidden_size)
+            self.ln1 = torch.nn.LayerNorm(hidden_size)
+            self.linear2 = torch.nn.Linear(hidden_size, hidden_size)
+            self.ln2 = torch.nn.LayerNorm(hidden_size)
+            self.bias = torch.nn.Parameter(torch.zeros(hidden_size))
+
+        def forward(self, x):
+            h = self.ln1(torch.nn.functional.relu(self.linear1(x)))
+            h = torch.nn.functional.relu(self.linear2(x))
+            return self.ln2(x + h + self.bias)
+
 
 @require_torch
 class TrainerUtilsTest(unittest.TestCase):
@@ -117,3 +132,12 @@ class TrainerUtilsTest(unittest.TestCase):
         self.assertEqual(lengths[indices_process_0[0]], 50)
         # The indices should be a permutation of range(100)
         self.assertEqual(list(sorted(indices_process_0 + indices_process_1)), list(range(100)))
+
+    def test_get_parameter_names(self):
+        model = torch.nn.Sequential(TstLayer(128), torch.nn.ModuleList([TstLayer(128), TstLayer(128)]))
+        # fmt: off
+        self.assertEqual(
+            get_parameter_names(model, [torch.nn.LayerNorm]),
+            ['0.linear1.weight', '0.linear1.bias', '0.linear2.weight', '0.linear2.bias', '0.bias', '1.0.linear1.weight', '1.0.linear1.bias', '1.0.linear2.weight', '1.0.linear2.bias', '1.0.bias', '1.1.linear1.weight', '1.1.linear1.bias', '1.1.linear2.weight', '1.1.linear2.bias', '1.1.bias']
+        )
+        # fmt: on