add fn_call example and bugfix gen_keyword

examples/assistant_data_analysis.py

@@ -17,10 +17,10 @@ def app():
     # chat
     messages = []
     while True:
-        # query example: Help me draw a line chart to show the changes in stock prices
+        # query example: pd.head the file first and then help me draw a line chart to show the changes in stock prices
         query = input('user question: ')
         # file example: resource/stock_prices.csv
-        file = input('file url (press enter if no file): ')
+        file = input('file url (press enter if no file): ').strip()
         if not query:
             print('user question cannot be empty！')
             continue

examples/assistant_doctor.py

@@ -31,7 +31,7 @@ def app():
         # query example: 医生，可以帮我看看我是否健康吗？
         query = input('user question: ')
         # file example: https://pic4.zhimg.com/80/v2-2c8eedf3e12386fedcd5589cf5575717_720w.webp
-        file = input('file url (press enter if no file): ')
+        file = input('file url (press enter if no file): ').strip()
         if not query:
             print('user question cannot be empty！')
             continue

examples/assistant_growing_girl.py

@@ -23,7 +23,7 @@ def app():
         # query example: 请开始创作！
         query = input('user question: ')
         # file example: resource/growing_girl.pdf
-        file = input('file url (press enter if no file): ')
+        file = input('file url (press enter if no file): ').strip()
         if not query:
             print('user question cannot be empty！')
             continue

examples/assistant_weather_bot.py

@@ -24,7 +24,7 @@ def app():
         # query example: 海淀区天气
         query = input('user question: ')
         # file example: resource/poem.pdf
-        file = input('file url (press enter if no file): ')
+        file = input('file url (press enter if no file): ').strip()
         if not query:
             print('user question cannot be empty！')
             continue

examples/function_calling.py

@@ -0,0 +1,114 @@
+# Reference: https://platform.openai.com/docs/guides/function-calling
+import json
+
+from qwen_agent.llm import get_chat_model
+
+
+# Example dummy function hard coded to return the same weather
+# In production, this could be your backend API or an external API
+def get_current_weather(location, unit='fahrenheit'):
+    """Get the current weather in a given location"""
+    if 'tokyo' in location.lower():
+        return json.dumps({
+            'location': 'Tokyo',
+            'temperature': '10',
+            'unit': 'celsius'
+        })
+    elif 'san francisco' in location.lower():
+        return json.dumps({
+            'location': 'San Francisco',
+            'temperature': '72',
+            'unit': 'fahrenheit'
+        })
+    elif 'paris' in location.lower():
+        return json.dumps({
+            'location': 'Paris',
+            'temperature': '22',
+            'unit': 'celsius'
+        })
+    else:
+        return json.dumps({'location': location, 'temperature': 'unknown'})
+
+
+def run_conversation():
+    llm = get_chat_model({
+        # Use the model service provided by DashScope:
+        'model': 'qwen-max',
+        'model_server': 'dashscope',
+        # 'api_key': 'YOUR_DASHSCOPE_API_KEY',
+        # It will use the `DASHSCOPE_API_KEY' environment variable if 'api_key' is not set.
+
+        # Use your own model service compatible with OpenAI API:
+        # 'model': 'Qwen/Qwen1.5-72B-Chat',
+        # 'model_server': 'http://localhost:8000/v1',  # api_base
+        # 'api_key': 'EMPTY',
+    })
+
+    # Step 1: send the conversation and available functions to the model
+    messages = [{
+        'role': 'user',
+        'content': "What's the weather like in San Francisco?"
+    }]
+    functions = [{
+        'name': 'get_current_weather',
+        'description': 'Get the current weather in a given location',
+        'parameters': {
+            'type': 'object',
+            'properties': {
+                'location': {
+                    'type': 'string',
+                    'description':
+                    'The city and state, e.g. San Francisco, CA',
+                },
+                'unit': {
+                    'type': 'string',
+                    'enum': ['celsius', 'fahrenheit']
+                },
+            },
+            'required': ['location'],
+        },
+    }]
+
+    print('# Assistant Response 1:')
+    responses = llm.chat(messages=messages, functions=functions, stream=False)
+    print(responses)
+
+    messages.extend(responses)  # extend conversation with assistant's reply
+
+    # Step 2: check if the model wanted to call a function
+    last_response = messages[-1]
+    if last_response.get('function_call', None):
+
+        # Step 3: call the function
+        # Note: the JSON response may not always be valid; be sure to handle errors
+        available_functions = {
+            'get_current_weather': get_current_weather,
+        }  # only one function in this example, but you can have multiple
+        function_name = last_response['function_call']['name']
+        function_to_call = available_functions[function_name]
+        function_args = json.loads(last_response['function_call']['arguments'])
+        function_response = function_to_call(
+            location=function_args.get('location'),
+            unit=function_args.get('unit'),
+        )
+        print('# Function Response:')
+        print(function_response)
+
+        # Step 4: send the info for each function call and function response to the model
+        messages.append({
+            'role': 'function',
+            'name': function_name,
+            'content': function_response,
+        })  # extend conversation with function response
+
+        print('# Assistant Response 2:')
+        responses = llm.chat(
+            messages=messages,
+            functions=functions,
+            stream=False,
+        )  # get a new response from the model where it can see the function response
+        print(responses)
+
+
+if __name__ == '__main__':
+    run_conversation()

examples/multi_agent_router.py

@@ -40,7 +40,7 @@ def app():
         # image example: https://img01.sc115.com/uploads/sc/jpgs/1505/apic11540_sc115.com.jpg
         image = input('image url (press enter if no image): ')
         # file example: resource/poem.pdf
-        file = input('file url (press enter if no file): ')
+        file = input('file url (press enter if no file): ').strip()
         if not query:
             print('user question cannot be empty！')
             continue

examples/visual_storytelling.py

@@ -66,7 +66,7 @@ def app():
     while True:
         query = input('user question: ')
         # image example: https://img01.sc115.com/uploads3/sc/vector/201809/51413-20180914205509.jpg
-        image = input('image url: ')
+        image = input('image url: ').strip()
 
         if not image:
             print('image cannot be empty！')

qwen_agent/agent.py

@@ -32,8 +32,7 @@ def __init__(self,
         :param kwargs: other potential parameters
         """
         if isinstance(llm, dict):
-            self.llm_config = llm
-            self.llm = get_chat_model(self.llm_config)
+            self.llm = get_chat_model(llm)
         else:
             self.llm = llm
 

qwen_agent/agents/assistant.py

@@ -1,8 +1,10 @@
+import copy
 from typing import Dict, Iterator, List, Optional, Union
 
 from qwen_agent import Agent
 from qwen_agent.llm import BaseChatModel
-from qwen_agent.llm.schema import CONTENT, DEFAULT_SYSTEM_MESSAGE, ROLE, SYSTEM
+from qwen_agent.llm.schema import (CONTENT, DEFAULT_SYSTEM_MESSAGE, FUNCTION,
+                                   ROLE, SYSTEM, Message)
 from qwen_agent.log import logger
 from qwen_agent.memory import Memory
 from qwen_agent.utils.utils import format_knowledge_to_source_and_content
@@ -48,11 +50,12 @@ def __init__(self,
         self.mem = Memory(llm=self.llm, files=files)
 
     def _run(self,
-             messages: List[Dict],
+             messages: List[Message],
              lang: str = 'zh',
              max_ref_token: int = 4000,
              **kwargs) -> Iterator[List[Dict]]:
-        system_prompt = ''
+        messages = copy.deepcopy(messages)
+        knowledge_prompt = ''
 
         # retrieval knowledge from files
         *_, last = self.mem.run(messages=messages, max_ref_token=max_ref_token)
@@ -64,13 +67,13 @@ def _run(self,
             for k in knowledge:
                 snippets.append(KNOWLEDGE_SNIPPET[lang].format(
                     source=k['source'], content=k['content']))
-            system_prompt += KNOWLEDGE_TEMPLATE[lang].format(
+            knowledge_prompt += KNOWLEDGE_TEMPLATE[lang].format(
                 knowledge='\n\n'.join(snippets))
 
         if messages[0][ROLE] == SYSTEM:
-            messages[0][CONTENT] += system_prompt
+            messages[0][CONTENT] += knowledge_prompt
         else:
-            messages.insert(0, {ROLE: SYSTEM, CONTENT: system_prompt})
+            messages.insert(0, Message(role=SYSTEM, content=knowledge_prompt))
 
         max_turn = 5
         response = []
@@ -81,19 +84,19 @@ def _run(self,
                 functions=[
                     func.function for func in self.function_map.values()
                 ])
-            output = []
+            output: List[Message] = []
             for output in output_stream:
                 yield response + output
             response.extend(output)
             messages.extend(output)
             use_tool, action, action_input, _ = self._detect_tool(response[-1])
             if use_tool:
                 observation = self._call_tool(action, action_input)
-                fn_msg = {
-                    'role': 'function',
-                    'name': action,
-                    'content': observation,
-                }
+                fn_msg = Message(
+                    role=FUNCTION,
+                    name=action,
+                    content=observation,
+                )
                 messages.append(fn_msg)
                 response.append(fn_msg)
                 yield response

qwen_agent/llm/base.py

@@ -83,9 +83,9 @@ def chat(
                                                stream=stream,
                                                delta_stream=delta_stream)
             if isinstance(output, list):
-                output = self._postprocess_messages_for_func_call(output)
+                output = self._postprocess_messages_for_fn_call(output)
             else:
-                output = self._postprocess_messages_iterator_for_func_call(
+                output = self._postprocess_messages_iterator_for_fn_call(
                     output)
         else:
             messages = self._preprocess_messages(messages)
@@ -108,7 +108,7 @@ def _chat_with_functions(
         delta_stream: bool = False
     ) -> Union[List[Message], Iterator[List[Message]]]:
 
-        messages = self._prepend_tool_message(messages, functions)
+        messages = self._prepend_fn_call_system(messages, functions)
         messages = self._preprocess_messages(messages)
 
         if messages and messages[-1][ROLE] == ASSISTANT:
@@ -147,8 +147,9 @@ def _chat_no_stream(
     ) -> List[Message]:
         raise NotImplementedError
 
-    def _prepend_tool_message(
-            self, messages: List[Message],
+    @staticmethod
+    def _prepend_fn_call_system(
+            messages: List[Message],
             functions: Optional[List[Dict]]) -> List[Message]:
         # prepend tool react prompt
         tool_desc_template = FN_CALL_TEMPLATE['en']
@@ -173,19 +174,19 @@ def _prepend_tool_message(
     @staticmethod
     def _format_as_multimodal_messages(
             messages: List[Message]) -> List[Message]:
-        new_messages = []
+        multimodal_messages = []
         for msg in messages:
             role = msg[ROLE]
             assert role in (USER, ASSISTANT, SYSTEM, FUNCTION)
             if role == FUNCTION:
-                new_messages.append(msg)
+                multimodal_messages.append(msg)
                 continue
 
             content = []
-            if isinstance(msg[CONTENT], str):
+            if isinstance(msg[CONTENT], str):  # if text content
                 if msg[CONTENT]:
                     content = [ContentItem(text=msg[CONTENT])]
-            elif isinstance(msg[CONTENT], list):
+            elif isinstance(msg[CONTENT], list):  # if multimodal content
                 files = []
                 for item in msg[CONTENT]:
                     for k, v in item.model_dump().items():
@@ -195,7 +196,7 @@ def _format_as_multimodal_messages(
                             content.append(item)
                         if k in ('file', 'image'):
                             files.append(v)
-                if files:
+                if (msg[ROLE] in (SYSTEM, USER)) and files:
                     has_zh = has_chinese_chars(content)
                     upload = []
                     for f in [get_basename_from_url(f) for f in files]:
@@ -211,20 +212,20 @@ def _format_as_multimodal_messages(
                                 upload.append(f'[file]({f})')
                     upload = ' '.join(upload)
                     if has_zh:
-                        upload = f'（上传了 {upload}）'
+                        upload = f'（上传了 {upload}）\n\n'
                     else:
-                        upload = f'(Uploaded {upload})'
+                        upload = f'(Uploaded {upload})\n\n'
                     content = [ContentItem(text=upload)
                                ] + content  # insert a text
             else:
                 raise TypeError
 
-            new_messages.append(
+            multimodal_messages.append(
                 Message(role=role,
                         content=content,
                         function_call=msg.function_call))
 
-        return new_messages
+        return multimodal_messages
 
     def _preprocess_messages(self, messages: List[Message]) -> List[Message]:
         """
@@ -274,7 +275,7 @@ def _preprocess_messages(self, messages: List[Message]) -> List[Message]:
                     break
         return new_messages
 
-    def _postprocess_messages_for_func_call(
+    def _postprocess_messages_for_fn_call(
             self, messages: List[Message]) -> List[Message]:
         """
         If the model calls function by built-in function call template, convert and display it in function_call format in return.
@@ -371,11 +372,11 @@ def _postprocess_messages_for_func_call(
                                                 new_content))  # no func call
         return new_messages
 
-    def _postprocess_messages_iterator_for_func_call(
+    def _postprocess_messages_iterator_for_fn_call(
             self,
             messages: Iterator[List[Message]]) -> Iterator[List[Message]]:
         for m in messages:
-            yield self._postprocess_messages_for_func_call(m)
+            yield self._postprocess_messages_for_fn_call(m)
 
     def _convert_messages_to_target_type(
             self, messages: List[Message],