Create handler.py

handler.py

@@ -0,0 +1,25 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+
+class ModelHandler:
+    def __init__(self):
+        self.tokenizer = None
+        self.model = None
+
+    def initialize(self, model_dir):
+        # Load the tokenizer and model
+        self.tokenizer = AutoTokenizer.from_pretrained(model_dir)
+        self.model = AutoModelForCausalLM.from_pretrained(model_dir)
+
+    def preprocess(self, inputs):
+        # Preprocess the input prompt
+        return self.tokenizer(inputs, return_tensors="pt", padding=True)
+
+    def inference(self, inputs):
+        # Generate text from the model
+        input_ids = inputs["input_ids"]
+        outputs = self.model.generate(input_ids, max_length=200, temperature=0.7)
+        return self.tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+    def postprocess(self, outputs):
+        return {"generated_text": outputs}