Create custom_model.py

custom_model.py

@@ -0,0 +1,35 @@
+# custom_model.py
+
+from transformers import PreTrainedModel, PretrainedConfig
+import torch
+import torch.nn as nn
+
+class CustomConfig(PretrainedConfig):
+    model_type = "custom_model"
+
+    def __init__(self, vocab_size=30522, hidden_size=768, num_hidden_layers=12, num_attention_heads=12, num_labels=2, **kwargs):
+        super().__init__(**kwargs)
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.num_labels = num_labels
+
+class CustomModel(PreTrainedModel):
+    config_class = CustomConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.embedding = nn.Embedding(config.vocab_size, config.hidden_size)
+        self.layers = nn.ModuleList([nn.TransformerEncoderLayer(d_model=config.hidden_size, nhead=config.num_attention_heads) for _ in range(config.num_hidden_layers)])
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+        
+        self.init_weights()
+
+    def forward(self, input_ids):
+        embeddings = self.embedding(input_ids)
+        x = embeddings
+        for layer in self.layers:
+            x = layer(x)
+        logits = self.classifier(x.mean(dim=1))  # Example: taking the mean of the output as input to the classifier
+        return logits