Validate llm_config passed to ConversableAgent (issue #1522) (#1654)

* Validate llm_config passed to ConversableAgent

Based on #1522, this commit implements the additional validation checks in
`ConversableAgent.`

Add the following validation and `raise ValueError` if:

 - The `llm_config` is `None`.
 - The `llm_config` is valid, but `config_list` is missing or lacks elements.
 - The `config_list` is valid, but no `model` is specified.

The rest of the changes are code churn to adjust or add the test cases.

* Validate llm_config passed to ConversableAgent

Based on #1522, this commit implements the additional validation checks in
`ConversableAgent.`

Add the following validation and `raise ValueError` if:

 - The `llm_config` is `None` (validated in `ConversableAgent`).
 - The `llm_config` has no `model` specified and `config_list` is empty
   (validated in `OpenAIWrapper`).
 - The `config_list` has at least one entry, but not all the entries have
   the `model` is specified (validated in `OpenAIWrapper`).

The rest of the changes are code churn to adjust or add the test cases.

* Validate llm_config passed to ConversableAgent

Based on #1522, this commit implements the additional validation checks in
`ConversableAgent.`

Add the following validation and `raise ValueError` if:

 - The `llm_config` is `None` (validated in `ConversableAgent`).
 - The `llm_config` has no `model` specified and `config_list` is empty
   (validated in `OpenAIWrapper`).
 - The `config_list` has at least one entry, but not all the entries have
   the `model` is specified (validated in `OpenAIWrapper`).

The rest of the changes are code churn to adjust or add the test cases.

* Validate llm_config passed to ConversableAgent

Based on #1522, this commit implements the additional validation checks in
`ConversableAgent.`

Add the following validation and `raise ValueError` if:

 - The `llm_config` is `None` (validated in `ConversableAgent`).
 - The `llm_config` has no `model` specified and `config_list` is empty
   (validated in `OpenAIWrapper`).
 - The `config_list` has at least one entry, but not all the entries have
   the `model` is specified (validated in `OpenAIWrapper`).

The rest of the changes are code churn to adjust or add the test cases.

* Validate llm_config passed to ConversableAgent

Based on #1522, this commit implements the additional validation checks in
`ConversableAgent.`

Add the following validation and `raise ValueError` if:

 - The `llm_config` is `None` (validated in `ConversableAgent`).
 - The `llm_config` has no `model` specified and `config_list` is empty
   (validated in `OpenAIWrapper`).
 - The `config_list` has at least one entry, but not all the entries have
   the `model` is specified (validated in `OpenAIWrapper`).

The rest of the changes are code churn to adjust or add the test cases.

* Fix the test_web_surfer issue

For anyone reading this: you need to `pip install markdownify` for the
`import WebSurferAgent` to succeed. That is needed to run the
`test_web_surfer.py` locally.

Test logic needs `llm_config` that is not `None` and that is not
`False`.

Let us pray that this works as part of GitHub actions ...

* One more fix for llm_config validation contract
This commit is contained in:
Gunnar Kudrjavets 2024-02-14 16:54:31 -08:00 committed by GitHub
parent 4ccff54dbe
commit d677b47c44
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 53 additions and 26 deletions

View File

@ -46,7 +46,7 @@ Reply "TERMINATE" in the end when everything is done.
name (str): agent name.
system_message (str): system message for the ChatCompletion inference.
Please override this attribute if you want to reprogram the agent.
llm_config (dict): llm inference configuration.
llm_config (dict or False or None): llm inference configuration.
Please refer to [OpenAIWrapper.create](/docs/reference/oai/client#create)
for available options.
is_termination_msg (function): a function that takes a message in the form of a dictionary

View File

@ -80,7 +80,7 @@ class ConversableAgent(LLMAgent):
function_map: Optional[Dict[str, Callable]] = None,
code_execution_config: Union[Dict, Literal[False]] = False,
llm_config: Optional[Union[Dict, Literal[False]]] = None,
default_auto_reply: Optional[Union[str, Dict, None]] = "",
default_auto_reply: Union[str, Dict] = "",
description: Optional[str] = None,
):
"""
@ -118,11 +118,11 @@ class ConversableAgent(LLMAgent):
- timeout (Optional, int): The maximum execution time in seconds.
- last_n_messages (Experimental, int or str): The number of messages to look back for code execution.
If set to 'auto', it will scan backwards through all messages arriving since the agent last spoke, which is typically the last time execution was attempted. (Default: auto)
llm_config (dict or False): llm inference configuration.
llm_config (dict or False or None): llm inference configuration.
Please refer to [OpenAIWrapper.create](/docs/reference/oai/client#create)
for available options.
To disable llm-based auto reply, set to False.
default_auto_reply (str or dict or None): default auto reply when no code execution or llm-based reply is generated.
default_auto_reply (str or dict): default auto reply when no code execution or llm-based reply is generated.
description (str): a short description of the agent. This description is used by other agents
(e.g. the GroupChatManager) to decide when to call upon this agent. (Default: system_message)
"""
@ -144,6 +144,11 @@ class ConversableAgent(LLMAgent):
self.llm_config = self.DEFAULT_CONFIG.copy()
if isinstance(llm_config, dict):
self.llm_config.update(llm_config)
# We still have a default `llm_config` because the user didn't
# specify anything. This won't work, so raise an error to avoid
# an obscure message from the OpenAI service.
if self.llm_config == {}:
raise ValueError("Please specify the value for 'llm_config'.")
self.client = OpenAIWrapper(**self.llm_config)
if logging_enabled():

View File

@ -361,8 +361,13 @@ class OpenAIWrapper:
if logging_enabled():
log_new_wrapper(self, locals())
openai_config, extra_kwargs = self._separate_openai_config(base_config)
# This *may* work if the `llm_config` has specified the `model` attribute,
# so just warn here.
if type(config_list) is list and len(config_list) == 0:
logger.warning("openai client was provided with an empty config_list, which may not be intended.")
logger.warning("OpenAI client was provided with an empty config_list, which may not be intended.")
# If the `llm_config` has no `model` then the call will fail. Abort now.
if "model" not in extra_kwargs:
raise ValueError("Please specify a value for the 'model' in 'llm_config'.")
self._clients: List[ModelClient] = []
self._config_list: List[Dict[str, Any]] = []
@ -370,6 +375,13 @@ class OpenAIWrapper:
if config_list:
config_list = [config.copy() for config in config_list] # make a copy before modifying
for config in config_list:
# We require that each element of `config_list` has a non-empty value
# for `model` specified unless `extra_kwargs` contains "model".
model = None
if "model" in config:
model = config["model"]
if "model" not in extra_kwargs and (model is None or len(model) == 0):
raise ValueError("Please specify a non-empty 'model' value for every item in 'config_list'.")
self._register_default_client(config, openai_config) # could modify the config
self._config_list.append(
{**extra_kwargs, **{k: v for k, v in config.items() if k not in self.openai_kwargs}}

View File

@ -51,7 +51,9 @@ def test_web_surfer() -> None:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
page_size = 4096
web_surfer = WebSurferAgent(
"web_surfer", llm_config={"config_list": []}, browser_config={"viewport_size": page_size}
"web_surfer",
llm_config={"model": "gpt-4", "config_list": []},
browser_config={"viewport_size": page_size},
)
# Sneak a peak at the function map, allowing us to call the functions for testing here

View File

@ -474,7 +474,7 @@ async def test_a_generate_reply_raises_on_messages_and_sender_none(conversable_a
def test_update_function_signature_and_register_functions() -> None:
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent = ConversableAgent(name="agent", llm_config={})
agent = ConversableAgent(name="agent", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
def exec_python(cell: str) -> None:
pass
@ -618,9 +618,9 @@ def get_origin(d: Dict[str, Callable[..., Any]]) -> Dict[str, Callable[..., Any]
def test_register_for_llm():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent3 = ConversableAgent(name="agent3", llm_config={"config_list": []})
agent2 = ConversableAgent(name="agent2", llm_config={"config_list": []})
agent1 = ConversableAgent(name="agent1", llm_config={"config_list": []})
agent3 = ConversableAgent(name="agent3", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
agent2 = ConversableAgent(name="agent2", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
agent1 = ConversableAgent(name="agent1", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
@agent3.register_for_llm()
@agent2.register_for_llm(name="python")
@ -691,9 +691,9 @@ def test_register_for_llm():
def test_register_for_llm_api_style_function():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent3 = ConversableAgent(name="agent3", llm_config={"config_list": []})
agent2 = ConversableAgent(name="agent2", llm_config={"config_list": []})
agent1 = ConversableAgent(name="agent1", llm_config={"config_list": []})
agent3 = ConversableAgent(name="agent3", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
agent2 = ConversableAgent(name="agent2", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
agent1 = ConversableAgent(name="agent1", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
@agent3.register_for_llm(api_style="function")
@agent2.register_for_llm(name="python", api_style="function")
@ -762,7 +762,7 @@ def test_register_for_llm_api_style_function():
def test_register_for_llm_without_description():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent = ConversableAgent(name="agent", llm_config={})
agent = ConversableAgent(name="agent", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
with pytest.raises(ValueError) as e:
@ -774,25 +774,33 @@ def test_register_for_llm_without_description():
def test_register_for_llm_without_LLM():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent = ConversableAgent(name="agent", llm_config=None)
agent.llm_config = None
assert agent.llm_config is None
try:
ConversableAgent(name="agent", llm_config=None)
assert False, "Expected ConversableAgent to throw ValueError."
except ValueError as e:
assert e.args[0] == "Please specify the value for 'llm_config'."
with pytest.raises(RuntimeError) as e:
@agent.register_for_llm(description="run cell in ipython and return the execution result.")
def exec_python(cell: Annotated[str, "Valid Python cell to execute."]) -> str:
pass
def test_register_for_llm_without_configuration():
try:
ConversableAgent(name="agent", llm_config={"config_list": []})
assert False, "Expected ConversableAgent to throw ValueError."
except ValueError as e:
assert e.args[0] == "Please specify a value for the 'model' in 'llm_config'."
assert e.value.args[0] == "LLM config must be setup before registering a function for LLM."
def test_register_for_llm_without_model_name():
try:
ConversableAgent(name="agent", llm_config={"config_list": [{"model": "", "api_key": ""}]})
assert False, "Expected ConversableAgent to throw ValueError."
except ValueError as e:
assert e.args[0] == "Please specify a non-empty 'model' value for every item in 'config_list'."
def test_register_for_execution():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent = ConversableAgent(name="agent", llm_config={"config_list": []})
agent = ConversableAgent(name="agent", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
user_proxy_1 = UserProxyAgent(name="user_proxy_1")
user_proxy_2 = UserProxyAgent(name="user_proxy_2")
@ -827,7 +835,7 @@ def test_register_for_execution():
def test_register_functions():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("OPENAI_API_KEY", MOCK_OPEN_AI_API_KEY)
agent = ConversableAgent(name="agent", llm_config={"config_list": []})
agent = ConversableAgent(name="agent", llm_config={"config_list": [{"model": "gpt-4", "api_key": ""}]})
user_proxy = UserProxyAgent(name="user_proxy")
def exec_python(cell: Annotated[str, "Valid Python cell to execute."]) -> str: