Custom Runtime Logger <> FileLogger (#2596)

* added logger param for custom logger support

* added FileLogger

* bump: spell check

* bump: import error

* added more log functionalites

* bump: builtin logger for FileLogger

* type check and instance level logger

* tests added for the fileLogger

* formatting bump

* updated tests and removed time formatting

* separate module for the filelogger

* update file logger test

* added the FileLogger into the notebook

* bump json decode error

* updated requested changes

* Updated tests with AutoGen agents

* bump file

* bump: logger accessed before intializedsolved

* Updated notebook to guide with a filename

* added thread_id to the FileLogger

* bump type check in tests

* Updated thread_id for each log event

* Updated thread_id for each log event

* Updated with tempfile

* bump: str cleanup

* skipping-windows tests

---------

Co-authored-by: Chi Wang <wang.chi@microsoft.com>
This commit is contained in:
HRUSHIKESH DOKALA 2024-05-16 11:24:05 +05:30 committed by GitHub
parent cd44932347
commit d461100bfa
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 451 additions and 10 deletions

View File

@ -0,0 +1,211 @@
from __future__ import annotations
import json
import logging
import os
import threading
import uuid
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union
from openai import AzureOpenAI, OpenAI
from openai.types.chat import ChatCompletion
from autogen.logger.base_logger import BaseLogger
from autogen.logger.logger_utils import get_current_ts, to_dict
from .base_logger import LLMConfig
if TYPE_CHECKING:
from autogen import Agent, ConversableAgent, OpenAIWrapper
logger = logging.getLogger(__name__)
class FileLogger(BaseLogger):
def __init__(self, config: Dict[str, Any]):
self.config = config
self.session_id = str(uuid.uuid4())
curr_dir = os.getcwd()
self.log_dir = os.path.join(curr_dir, "autogen_logs")
os.makedirs(self.log_dir, exist_ok=True)
self.log_file = os.path.join(self.log_dir, self.config.get("filename", "runtime.log"))
try:
with open(self.log_file, "a"):
pass
except Exception as e:
logger.error(f"[file_logger] Failed to create logging file: {e}")
self.logger = logging.getLogger(__name__)
self.logger.setLevel(logging.INFO)
file_handler = logging.FileHandler(self.log_file)
self.logger.addHandler(file_handler)
def start(self) -> str:
"""Start the logger and return the session_id."""
try:
self.logger.info(f"Started new session with Session ID: {self.session_id}")
except Exception as e:
logger.error(f"[file_logger] Failed to create logging file: {e}")
finally:
return self.session_id
def log_chat_completion(
self,
invocation_id: uuid.UUID,
client_id: int,
wrapper_id: int,
request: Dict[str, Union[float, str, List[Dict[str, str]]]],
response: Union[str, ChatCompletion],
is_cached: int,
cost: float,
start_time: str,
) -> None:
"""
Log a chat completion.
"""
thread_id = threading.get_ident()
try:
log_data = json.dumps(
{
"invocation_id": str(invocation_id),
"client_id": client_id,
"wrapper_id": wrapper_id,
"request": to_dict(request),
"response": str(response),
"is_cached": is_cached,
"cost": cost,
"start_time": start_time,
"end_time": get_current_ts(),
"thread_id": thread_id,
}
)
self.logger.info(log_data)
except Exception as e:
self.logger.error(f"[file_logger] Failed to log chat completion: {e}")
def log_new_agent(self, agent: ConversableAgent, init_args: Dict[str, Any] = {}) -> None:
"""
Log a new agent instance.
"""
thread_id = threading.get_ident()
try:
log_data = json.dumps(
{
"id": id(agent),
"agent_name": agent.name if hasattr(agent, "name") and agent.name is not None else "",
"wrapper_id": to_dict(
agent.client.wrapper_id if hasattr(agent, "client") and agent.client is not None else ""
),
"session_id": self.session_id,
"current_time": get_current_ts(),
"agent_type": type(agent).__name__,
"args": to_dict(init_args),
"thread_id": thread_id,
}
)
self.logger.info(log_data)
except Exception as e:
self.logger.error(f"[file_logger] Failed to log new agent: {e}")
def log_event(self, source: Union[str, Agent], name: str, **kwargs: Dict[str, Any]) -> None:
"""
Log an event from an agent or a string source.
"""
from autogen import Agent
# This takes an object o as input and returns a string. If the object o cannot be serialized, instead of raising an error,
# it returns a string indicating that the object is non-serializable, along with its type's qualified name obtained using __qualname__.
json_args = json.dumps(kwargs, default=lambda o: f"<<non-serializable: {type(o).__qualname__}>>")
thread_id = threading.get_ident()
if isinstance(source, Agent):
try:
log_data = json.dumps(
{
"source_id": id(source),
"source_name": str(source.name) if hasattr(source, "name") else source,
"event_name": name,
"agent_module": source.__module__,
"agent_class": source.__class__.__name__,
"json_state": json_args,
"timestamp": get_current_ts(),
"thread_id": thread_id,
}
)
self.logger.info(log_data)
except Exception as e:
self.logger.error(f"[file_logger] Failed to log event {e}")
else:
try:
log_data = json.dumps(
{
"source_id": id(source),
"source_name": str(source.name) if hasattr(source, "name") else source,
"event_name": name,
"json_state": json_args,
"timestamp": get_current_ts(),
"thread_id": thread_id,
}
)
self.logger.info(log_data)
except Exception as e:
self.logger.error(f"[file_logger] Failed to log event {e}")
def log_new_wrapper(
self, wrapper: OpenAIWrapper, init_args: Dict[str, Union[LLMConfig, List[LLMConfig]]] = {}
) -> None:
"""
Log a new wrapper instance.
"""
thread_id = threading.get_ident()
try:
log_data = json.dumps(
{
"wrapper_id": id(wrapper),
"session_id": self.session_id,
"json_state": json.dumps(init_args),
"timestamp": get_current_ts(),
"thread_id": thread_id,
}
)
self.logger.info(log_data)
except Exception as e:
self.logger.error(f"[file_logger] Failed to log event {e}")
def log_new_client(self, client: AzureOpenAI | OpenAI, wrapper: OpenAIWrapper, init_args: Dict[str, Any]) -> None:
"""
Log a new client instance.
"""
thread_id = threading.get_ident()
try:
log_data = json.dumps(
{
"client_id": id(client),
"wrapper_id": id(wrapper),
"session_id": self.session_id,
"class": type(client).__name__,
"json_state": json.dumps(init_args),
"timestamp": get_current_ts(),
"thread_id": thread_id,
}
)
self.logger.info(log_data)
except Exception as e:
self.logger.error(f"[file_logger] Failed to log event {e}")
def get_connection(self) -> None:
"""Method is intentionally left blank because there is no specific connection needed for the FileLogger."""
pass
def stop(self) -> None:
"""Close the file handler and remove it from the logger."""
for handler in self.logger.handlers:
if isinstance(handler, logging.FileHandler):
handler.close()
self.logger.removeHandler(handler)

View File

@ -1,6 +1,7 @@
from typing import Any, Dict, Optional
from typing import Any, Dict, Literal, Optional
from autogen.logger.base_logger import BaseLogger
from autogen.logger.file_logger import FileLogger
from autogen.logger.sqlite_logger import SqliteLogger
__all__ = ("LoggerFactory",)
@ -8,11 +9,15 @@ __all__ = ("LoggerFactory",)
class LoggerFactory:
@staticmethod
def get_logger(logger_type: str = "sqlite", config: Optional[Dict[str, Any]] = None) -> BaseLogger:
def get_logger(
logger_type: Literal["sqlite", "file"] = "sqlite", config: Optional[Dict[str, Any]] = None
) -> BaseLogger:
if config is None:
config = {}
if logger_type == "sqlite":
return SqliteLogger(config)
elif logger_type == "file":
return FileLogger(config)
else:
raise ValueError(f"[logger_factory] Unknown logger type: {logger_type}")

View File

@ -3,12 +3,12 @@ from __future__ import annotations
import logging
import sqlite3
import uuid
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
from typing import TYPE_CHECKING, Any, Dict, List, Literal, Optional, Union
from openai import AzureOpenAI, OpenAI
from openai.types.chat import ChatCompletion
from autogen.logger.base_logger import LLMConfig
from autogen.logger.base_logger import BaseLogger, LLMConfig
from autogen.logger.logger_factory import LoggerFactory
if TYPE_CHECKING:
@ -20,10 +20,26 @@ autogen_logger = None
is_logging = False
def start(logger_type: str = "sqlite", config: Optional[Dict[str, Any]] = None) -> str:
def start(
logger: Optional[BaseLogger] = None,
logger_type: Literal["sqlite", "file"] = "sqlite",
config: Optional[Dict[str, Any]] = None,
) -> str:
"""
Start logging for the runtime.
Args:
logger (BaseLogger): A logger instance
logger_type (str): The type of logger to use (default: sqlite)
config (dict): Configuration for the logger
Returns:
session_id (str(uuid.uuid4)): a unique id for the logging session
"""
global autogen_logger
global is_logging
if logger:
autogen_logger = logger
else:
autogen_logger = LoggerFactory.get_logger(logger_type=logger_type, config=config)
try:

View File

@ -8,6 +8,10 @@
"\n",
"AutoGen offers utilities to log data for debugging and performance analysis. This notebook demonstrates how to use them. \n",
"\n",
"we log data in different modes:\n",
"- SQlite Database\n",
"- File \n",
"\n",
"In general, users can initiate logging by calling `autogen.runtime_logging.start()` and stop logging by calling `autogen.runtime_logging.stop()`"
]
},
@ -287,6 +291,82 @@
" + str(round(session_cost, 4))\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Log data in File mode\n",
"\n",
"By default, the log type is set to `sqlite` as shown above, but we introduced a new parameter for the `autogen.runtime_logging.start()`\n",
"\n",
"the `logger_type = \"file\"` will start to log data in the File mode."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Logging session ID: ed493ebf-d78e-49f0-b832-69557276d557\n",
"\u001b[33muser_proxy\u001b[0m (to assistant):\n",
"\n",
"What is the height of the Eiffel Tower? Only respond with the answer and terminate\n",
"\n",
"--------------------------------------------------------------------------------\n",
"\u001b[33massistant\u001b[0m (to user_proxy):\n",
"\n",
"The height of the Eiffel Tower is 330 meters.\n",
"TERMINATE\n",
"\n",
"--------------------------------------------------------------------------------\n"
]
}
],
"source": [
"\n",
"import pandas as pd\n",
"\n",
"import autogen\n",
"from autogen import AssistantAgent, UserProxyAgent\n",
"\n",
"# Setup API key. Add your own API key to config file or environment variable\n",
"llm_config = {\n",
" \"config_list\": autogen.config_list_from_json(\n",
" env_or_file=\"OAI_CONFIG_LIST\",\n",
" ),\n",
" \"temperature\": 0.9,\n",
"}\n",
"\n",
"# Start logging with logger_type and the filename to log to\n",
"logging_session_id = autogen.runtime_logging.start(logger_type=\"file\", config={\"filename\": \"runtime.log\"})\n",
"print(\"Logging session ID: \" + str(logging_session_id))\n",
"\n",
"# Create an agent workflow and run it\n",
"assistant = AssistantAgent(name=\"assistant\", llm_config=llm_config)\n",
"user_proxy = UserProxyAgent(\n",
" name=\"user_proxy\",\n",
" code_execution_config=False,\n",
" human_input_mode=\"NEVER\",\n",
" is_termination_msg=lambda msg: \"TERMINATE\" in msg[\"content\"],\n",
")\n",
"\n",
"user_proxy.initiate_chat(\n",
" assistant, message=\"What is the height of the Eiffel Tower? Only respond with the answer and terminate\"\n",
")\n",
"autogen.runtime_logging.stop()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This should create a `runtime.log` file in your current directory. "
]
}
],
"metadata": {
@ -312,7 +392,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.7"
"version": "3.9.13"
}
},
"nbformat": 4,

View File

@ -0,0 +1,127 @@
import json
import os
import sys
import tempfile
import uuid
import pytest
sys.path.append(os.path.join(os.path.dirname(__file__), ".."))
sys.path.append(os.path.join(os.path.dirname(__file__), "../.."))
from conftest import skip_openai # noqa: E402
from test_assistant_agent import KEY_LOC, OAI_CONFIG_LIST # noqa: E402
import autogen
import autogen.runtime_logging
from autogen.logger.file_logger import FileLogger
is_windows = sys.platform.startswith("win")
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
@pytest.fixture
def logger() -> FileLogger:
current_dir = os.path.dirname(os.path.abspath(__file__))
with tempfile.TemporaryDirectory(dir=current_dir) as temp_dir:
log_file = os.path.join(temp_dir, "test_log.log")
config = {"filename": log_file}
logger = FileLogger(config)
yield logger
logger.stop()
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
def test_start(logger: FileLogger):
session_id = logger.start()
assert isinstance(session_id, str)
assert len(session_id) == 36
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
def test_log_chat_completion(logger: FileLogger):
invocation_id = uuid.uuid4()
client_id = 123456789
wrapper_id = 987654321
request = {"messages": [{"content": "Test message", "role": "user"}]}
response = "Test response"
is_cached = 0
cost = 0.5
start_time = "2024-05-06 15:20:21.263231"
logger.log_chat_completion(invocation_id, client_id, wrapper_id, request, response, is_cached, cost, start_time)
with open(logger.log_file, "r") as f:
lines = f.readlines()
assert len(lines) == 1
log_data = json.loads(lines[0])
assert log_data["invocation_id"] == str(invocation_id)
assert log_data["client_id"] == client_id
assert log_data["wrapper_id"] == wrapper_id
assert log_data["response"] == response
assert log_data["is_cached"] == is_cached
assert log_data["cost"] == cost
assert log_data["start_time"] == start_time
assert isinstance(log_data["thread_id"], int)
class TestWrapper:
def __init__(self, init_args):
self.init_args = init_args
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
def test_log_new_agent(logger: FileLogger):
agent = autogen.UserProxyAgent(name="user_proxy", code_execution_config=False)
logger.log_new_agent(agent)
with open(logger.log_file, "r") as f:
lines = f.readlines()
log_data = json.loads(lines[0]) # the first line is the session id
assert log_data["agent_name"] == "user_proxy"
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
def test_log_event(logger: FileLogger):
source = autogen.AssistantAgent(name="TestAgent", code_execution_config=False)
name = "TestEvent"
kwargs = {"key": "value"}
logger.log_event(source, name, **kwargs)
with open(logger.log_file, "r") as f:
lines = f.readlines()
log_data = json.loads(lines[0])
assert log_data["source_name"] == "TestAgent"
assert log_data["event_name"] == name
assert log_data["json_state"] == json.dumps(kwargs)
assert isinstance(log_data["thread_id"], int)
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
def test_log_new_wrapper(logger: FileLogger):
wrapper = TestWrapper(init_args={"foo": "bar"})
logger.log_new_wrapper(wrapper, wrapper.init_args)
with open(logger.log_file, "r") as f:
lines = f.readlines()
log_data = json.loads(lines[0])
assert log_data["wrapper_id"] == id(wrapper)
assert log_data["json_state"] == json.dumps(wrapper.init_args)
assert isinstance(log_data["thread_id"], int)
@pytest.mark.skipif(is_windows, reason="Skipping file logging tests on Windows")
def test_log_new_client(logger: FileLogger):
client = autogen.UserProxyAgent(name="user_proxy", code_execution_config=False)
wrapper = TestWrapper(init_args={"foo": "bar"})
init_args = {"foo": "bar"}
logger.log_new_client(client, wrapper, init_args)
with open(logger.log_file, "r") as f:
lines = f.readlines()
log_data = json.loads(lines[0])
assert log_data["client_id"] == id(client)
assert log_data["wrapper_id"] == id(wrapper)
assert log_data["json_state"] == json.dumps(init_args)
assert isinstance(log_data["thread_id"], int)

View File

@ -6,14 +6,16 @@ import uuid
import pytest
import autogen
import autogen.runtime_logging
sys.path.append(os.path.join(os.path.dirname(__file__), ".."))
sys.path.append(os.path.join(os.path.dirname(__file__), "../.."))
from conftest import skip_openai # noqa: E402
from test_assistant_agent import KEY_LOC, OAI_CONFIG_LIST # noqa: E402
import autogen
import autogen.runtime_logging
TEACHER_MESSAGE = """
You are roleplaying a math teacher, and your job is to help your students with linear algebra.
Keep your explanations short.