Update README.md

README.md

@@ -65,15 +65,17 @@ See the [author's demo](https://huggingface.co/spaces/tonic/gaiaminimed)
 Use the code below to get started with the model.
 
 ```python
-from transformers import AutoTokenizer, AutoModelForCausalLM
+from transformers import AutoConfig, AutoTokenizer, AutoModelForCausalLM
+from peft import PeftModel, PeftConfig
 import torch
 import gradio as gr
 
 # Define the device
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
+
 # Use model IDs as variables
-base_model_id = "mistralai/Mistral-7B-v0.1"
+base_model_id = "tiiuae/falcon-7b-instruct"
 model_directory = "Tonic/GaiaMiniMed"
 
 # Instantiate the Tokenizer
@@ -82,13 +84,16 @@ tokenizer.pad_token = tokenizer.eos_token
 tokenizer.padding_side = 'left'
 
 
+
 # Load the GaiaMiniMed model with the specified configuration
-# Load the Peft model with a specific configuration
-peft_config = PeftConfig.from_pretrained("Tonic/GaiaMiniMed")
-peft_model = PeftModel.from_pretrained("Tonic/GaiaMiniMed", config=peft_config)
+# Specify the configuration class for the model
+model_config = AutoConfig.from_pretrained(base_model_id)
+# Load the PEFT model with the specified configuration
+peft_model = AutoModelForCausalLM.from_pretrained(base_model_id, config=model_config)
+peft_model = PeftModel.from_pretrained("Tonic/GaiaMiniMed")
+peft_model = PeftModel.from_pretrained(peft_model, "Tonic/GaiaMiniMed")
+
 
-# Now you can use peft_model without any NameError
-peft_model = peft_model.to_bettertransformer("tiiuae/falcon-7b-instruct")
 
 # Class to encapsulate the Falcon chatbot
 class FalconChatBot:
@@ -125,7 +130,7 @@ class FalconChatBot:
         input_ids = tokenizer.encode(conversation, return_tensors="pt", add_special_tokens=False)
 
         # Generate a response using the Falcon model
-        response = falcon_model.generate(input_ids, max_length=max_length, use_cache=True, early_stopping=True, bos_token_id=falcon_model.config.bos_token_id, eos_token_id=falcon_model.config.eos_token_id, pad_token_id=falcon_model.config.eos_token_id, temperature=0.1, do_sample=True)
+        response = falcon_model.generate(input_ids, max_length=max_length, use_cache=True, early_stopping=True, bos_token_id=falcon_model.config.bos_token_id, eos_token_id=falcon_model.config.eos_token_id, pad_token_id=falcon_model.config.eos_token_id, temperature=0.4, do_sample=True)
 
         # Decode the generated response to text
         response_text = tokenizer.decode(response[0], skip_special_tokens=True)