Upload MLP

config.json

@@ -0,0 +1,16 @@
+{
+  "_name_or_path": "zhihanyang/mlp",
+  "architectures": [
+    "MLP"
+  ],
+  "auto_map": {
+    "AutoConfig": "config.MLPConfig",
+    "AutoModelForMaskedLM": "model.MLP"
+  },
+  "hidden_dim": 32,
+  "input_dim": 16,
+  "model_type": "MLP",
+  "output_dim": 16,
+  "torch_dtype": "float32",
+  "transformers_version": "4.49.0"
+}

config.py

@@ -0,0 +1,16 @@
+import transformers
+
+
+class MLPConfig(transformers.PretrainedConfig):
+  model_type = 'MLP'
+
+  def __init__(
+    self,
+    input_dim: int = 16,
+    hidden_dim: int = 32,
+    output_dim: int = 16,
+    **kwargs):
+    super().__init__(**kwargs)
+    self.input_dim = input_dim
+    self.hidden_dim = hidden_dim
+    self.output_dim = output_dim

model.py

@@ -0,0 +1,32 @@
+import torch
+import torch.nn as nn
+import typing
+import transformers
+
+from .config import MLPConfig
+
+
+class Backbone(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.model = nn.Sequential(
+            nn.Linear(config.input_dim, config.hidden_dim),
+            nn.ReLU(),
+            nn.Linear(config.hidden_dim, config.hidden_dim),
+            nn.ReLU(),
+            nn.Linear(config.hidden_dim, config.output_dim)
+        )
+
+    def forward(self, x):
+        return self.model(x)
+
+
+class MLP(transformers.PreTrainedModel):
+  """HF-compatible model."""
+  config_class = MLPConfig
+  base_model_prefix = 'mlp'
+
+  def __init__(self, config: MLPConfig):
+    super().__init__(config)
+    self.config = config
+    self.backbone = Backbone(config)

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aba70e689e01bd5aec5926f6dbf032ae12b984380fa47684108b607047a9ef44
+size 9048