Update README.md

README.md

@@ -15,29 +15,38 @@ For more details, please see the accompanying paper.
 ### Training data
 The model is trained on 13k examples from the training subset of the [GeoCoDe dataset](https://github.com/EgoLaparra/geocode-data), where the input is a complex location reference and the center coordinates of each mentioned location and the output is the location's corresponding bounding box. 
 
-### Intended uses and limitations
-Due to data limitations, this model has been trained and evaluated for our task only in Mainstream American English.
+### Limitations
+Due to data limitations, this model has been trained and evaluated for our task only in Mainstream American English. 
 
 
 ### Usage
-We have included sample code below to use the model. For the system prompt and example prompts, please see the appendices in the accompanying paper. 
+The following code snippet illustrates how to use the model. For the system prompt we used and for example prompts, please see the appendices in the accompanying paper. 
 
-```
-from unsloth import FastLanguageModel
-import torch
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# Load model from Huggingface Hub
-model, tokenizer = FastLanguageModel.from_pretrained(
-    model_name = "tmasis/geocoding-complex-location-references",
-    max_seq_length = 2048,
-    load_in_4bit = True)
-FastLanguageModel.for_inference(model)
+model_name = "tmasis/geocoding-complex-location-references"
 
+# Load model and tokenizer from Huggingface Hub
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name = model_name,
+    torch_dtype = "auto",
+    device_map = "auto"
+)
+
+# Prepare model input
 messages = [{"role": "system", "content": <system_prompt>},
     {"role": "user", "content": <prompt>}]
-text = tokenizer.apply_chat_template(messages, tokenizer=False,
-    add_generation_prompt = True, enable_thinking = False)
-outputs = model.generate(**tokenizer(text, return_tensors="pt").to("cuda"),
+text = tokenizer.apply_chat_template(messages,
+    tokenize=False,
+    add_generation_prompt = True,
+    enable_thinking = False
+)
+
+# Conduct text generation
+outputs = model.generate(**tokenizer(text, return_tensors="pt").to(model.device),
     max_new_tokens=1024, temperature=0.7, top_p=0.8, top_k=20)
 response = tokenizer.batch_decode(outputs)[0]
+print(response)
 ```