Update README.md

README.md

@@ -17,13 +17,151 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
 
+Example query with prompt:
+
+````python
+import json
+
+import torch 
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from jinja2 import Template
+
+model_name = "plaguss/Llama-3.2-1B-Instruct-APIGen-FC-v0.1-notoken"
+model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto", trust_remote_code=True)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+SYSTEM_PROMPT = """
+You are an expert in composing functions. You are given a question and a set of possible functions. 
+Based on the question, you will need to make one or more function/tool calls to achieve the purpose. 
+If none of the functions can be used, point it out and refuse to answer. 
+If the given question lacks the parameters required by the function, also point it out.
+
+The output MUST strictly adhere to the following format, and NO other text MUST be included.
+The example format is as follows. Please make sure the parameter type is correct. If no function call is needed, please make the tool calls an empty list '[]'.
+```
+<tool_call>[
+{"name": "func_name1", "arguments": {"argument1": "value1", "argument2": "value2"}},
+... (more tool calls as required)
+]</tool_call>
+```
+""".strip()
+
+prompt = Template("""
+You have access to the following tools:
+<tools>{{ tools }}</tools>
+
+Please answer the following query:
+{{ query }}
+""".lstrip())
+
+get_weather_api = {
+    "name": "get_weather",
+    "description": "Get the current weather for a location",
+    "parameters": {
+        "type": "object",
+        "properties": {
+            "location": {
+                "type": "string",
+                "description": "The city and state, e.g. San Francisco, New York"
+            },
+            "unit": {
+                "type": "string",
+                "enum": ["celsius", "fahrenheit"],
+                "description": "The unit of temperature to return"
+            }
+        },
+        "required": ["location"]
+    }
+}
+
+search_api = {
+    "name": "search",
+    "description": "Search for information on the internet",
+    "parameters": {
+        "type": "object",
+        "properties": {
+            "query": {
+                "type": "string",
+                "description": "The search query, e.g. 'latest news on AI'"
+            }
+        },
+        "required": ["query"]
+    }
+}
+
+tools = [get_weather_api, search_api]
+
+query = "What's the weather like in New York in fahrenheit?"
+
+user_prompt = prompt.render(tools=json.dumps(tools), query=query)
+
+messages=[
+    {"role": "system", "content": SYSTEM_PROMPT},
+    { 'role': 'user', 'content': user_prompt}
+]
+inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
+
+outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
+result = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=False)
+
+pattern = r'<tool_call>(.*?)</tool_call>'
+import re
+matches = re.findall(pattern, result, re.DOTALL)
+response = json.loads(matches[0])
+# [{'name': 'get_weather', 'arguments': {'location': 'New York', 'unit': 'fahrenheit'}}]
+````
+
+Example response with no tools available
+
 ```python
-from transformers import pipeline
+tools = "[]"
+
+query = "What's the weather like in New York in fahrenheit?"
+
+user_prompt = prompt.render(tools=json.dumps(tools), query=query)
+
+messages=[
+    {"role": "system", "content": SYSTEM_PROMPT},
+    { 'role': 'user', 'content': user_prompt}
+]
+inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
+
+outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
+result = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True)
+# 'The query cannot be answered, no tools were provided.'
+```
+
+Example when a wrong tool is informed:
+
+```python
+cut_number = {
+  'type': 'function',
+  'function': {
+       'name': 'cut_number',
+       'description': 'Returns the value `number` if it is greater than or equal to `threshold`, otherwise returns the value `threshold`.',
+       'parameters': {
+          'type': 'object',
+          'properties': {'number': {'type': 'number', 'description': 'The number to compare.'}},
+      'required': ['number']
+      }
+  }
+}
+
+tools = [cut_number]
+
+query = "What's the weather like in New York in fahrenheit?"
+
+user_prompt = prompt.render(tools=json.dumps(tools), query=query)
+
+messages=[
+    {"role": "system", "content": SYSTEM_PROMPT},
+    { 'role': 'user', 'content': user_prompt}
+]
+inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
 
-question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="plaguss/Llama-3.2-1B-Instruct-APIGen-FC-v0.1-notoken", device="cuda")
-output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
-print(output["generated_text"])
+outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
+result = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True)
+# "The query cannot be answered with the provided tools. The query lacks the parameters required by the function. Please provide the parameters, and I'll be happy to assist."
 ```
 
 ## Training procedure