Update app.py

app.py

@@ -9,12 +9,28 @@ model = AutoModelForCausalLM.from_pretrained(
     return_dict=True,
     device_map="auto"
 )
-tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
+#tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
+
+mixtral_tokenizer = AutoTokenizer.from_pretrained(peft_model_id)
 
 # Load the Lora model
 model = PeftModel.from_pretrained(model, peft_model_id)
 
-
+def input_from_text(product, description):
+  return f"<s>[INST]Below is a product and description, please write a marketing email for this product.\n\n### Product:\n{product}\n### Description:\n{description}\n\n### Marketing Email:[/INST]"
+
+def make_inference(product, description):
+  inputs = mixtral_tokenizer(input_from_text(product, description), return_tensors="pt")
+
+  outputs = merged_model.generate(
+      **inputs,
+      max_new_tokens=150,
+      generation_kwargs={"repetition_penalty" : 1.7}
+      )
+  # print(mixtral_tokenizer.decode(outputs[0], skip_special_tokens=True))
+  result = mixtral_tokenizer.decode(outputs[0], skip_special_tokens=True).split("[/INST]")[1]
+  return result
+'''
 def make_inference(product_name, product_description):
     batch = tokenizer(
         f"### Product and Description:\n{product_name}: {product_description}\n\n### Ad:",
@@ -28,7 +44,7 @@ def make_inference(product_name, product_description):
 
     return tokenizer.decode(output_tokens[0], skip_special_tokens=True)
 
-
+'''
 if __name__ == "__main__":
     # make a gradio interface
     import gradio as gr