github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-07-04 12:18:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

Session to project (#6249)

Browse Source 
Sessions are being renamed to projects in the tracer
This commit is contained in:
Zander Chase 2023-06-23 01:11:01 -07:00 committed by GitHub
parent 9c09861946
commit b4fe7f3a09
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
7 changed files with 724 additions and 716 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
langchain/callbacks/manager.py
View File

@ -106,7 +106,7 @@ def wandb_tracing_enabled(
@contextmanager @contextmanager
def tracing_v2_enabled( def tracing_v2_enabled(
session_name: Optional[str] = None, project_name: Optional[str] = None,
*, *,
example_id: Optional[Union[str, UUID]] = None, example_id: Optional[Union[str, UUID]] = None,
) -> Generator[None, None, None]: ) -> Generator[None, None, None]:
@ -120,7 +120,7 @@ def tracing_v2_enabled(
example_id = UUID(example_id) example_id = UUID(example_id)
cb = LangChainTracer( cb = LangChainTracer(
example_id=example_id, example_id=example_id,
session_name=session_name, project_name=project_name,
) )
tracing_v2_callback_var.set(cb) tracing_v2_callback_var.set(cb)
yield yield
@ -131,12 +131,12 @@ def tracing_v2_enabled(
def trace_as_chain_group( def trace_as_chain_group(
group_name: str, group_name: str,
*, *,
session_name: Optional[str] = None, project_name: Optional[str] = None,
example_id: Optional[Union[str, UUID]] = None, example_id: Optional[Union[str, UUID]] = None,
) -> Generator[CallbackManager, None, None]: ) -> Generator[CallbackManager, None, None]:
"""Get a callback manager for a chain group in a context manager.""" """Get a callback manager for a chain group in a context manager."""
cb = LangChainTracer( cb = LangChainTracer(
session_name=session_name, project_name=project_name,
example_id=example_id, example_id=example_id,
) )
cm = CallbackManager.configure( cm = CallbackManager.configure(
@ -152,12 +152,12 @@ def trace_as_chain_group(
async def atrace_as_chain_group( async def atrace_as_chain_group(
group_name: str, group_name: str,
*, *,
session_name: Optional[str] = None, project_name: Optional[str] = None,
example_id: Optional[Union[str, UUID]] = None, example_id: Optional[Union[str, UUID]] = None,
) -> AsyncGenerator[AsyncCallbackManager, None]: ) -> AsyncGenerator[AsyncCallbackManager, None]:
"""Get a callback manager for a chain group in a context manager.""" """Get a callback manager for a chain group in a context manager."""
cb = LangChainTracer( cb = LangChainTracer(
session_name=session_name, project_name=project_name,
example_id=example_id, example_id=example_id,
) )
cm = AsyncCallbackManager.configure( cm = AsyncCallbackManager.configure(
@ -1039,10 +1039,10 @@ def _configure(
tracing_v2_enabled_ = ( tracing_v2_enabled_ = (
env_var_is_set("LANGCHAIN_TRACING_V2") or tracer_v2 is not None env_var_is_set("LANGCHAIN_TRACING_V2") or tracer_v2 is not None
) )
tracer_session = os.environ.get("LANGCHAIN_SESSION") tracer_project = os.environ.get(
"LANGCHAIN_PROJECT", os.environ.get("LANGCHAIN_SESSION", "default")
)
debug = _get_debug() debug = _get_debug()
if tracer_session is None:
tracer_session = "default"
if ( if (
verbose verbose
or debug or debug
@ -1072,7 +1072,7 @@ def _configure(
callback_manager.add_handler(tracer, True) callback_manager.add_handler(tracer, True)
else: else:
handler = LangChainTracerV1() handler = LangChainTracerV1()
handler.load_session(tracer_session) handler.load_session(tracer_project)
callback_manager.add_handler(handler, True) callback_manager.add_handler(handler, True)
if wandb_tracing_enabled_ and not any( if wandb_tracing_enabled_ and not any(
isinstance(handler, WandbTracer) for handler in callback_manager.handlers isinstance(handler, WandbTracer) for handler in callback_manager.handlers
@ -1090,7 +1090,7 @@ def _configure(
callback_manager.add_handler(tracer_v2, True) callback_manager.add_handler(tracer_v2, True)
else: else:
try: try:
handler = LangChainTracer(session_name=tracer_session) handler = LangChainTracer(project_name=tracer_project)
callback_manager.add_handler(handler, True) callback_manager.add_handler(handler, True)
except Exception as e: except Exception as e:
logger.warning( logger.warning(

8
langchain/callbacks/tracers/langchain.py
View File

@ -45,7 +45,7 @@ class LangChainTracer(BaseTracer):
def __init__( def __init__(
self, self,
example_id: Optional[Union[UUID, str]] = None, example_id: Optional[Union[UUID, str]] = None,
session_name: Optional[str] = None, project_name: Optional[str] = None,
client: Optional[LangChainPlusClient] = None, client: Optional[LangChainPlusClient] = None,
**kwargs: Any, **kwargs: Any,
) -> None: ) -> None:
@ -55,7 +55,9 @@ class LangChainTracer(BaseTracer):
self.example_id = ( self.example_id = (
UUID(example_id) if isinstance(example_id, str) else example_id UUID(example_id) if isinstance(example_id, str) else example_id
) )
self.session_name = session_name or os.getenv("LANGCHAIN_SESSION", "default") self.project_name = project_name or os.getenv(
"LANGCHAIN_PROJECT", os.getenv("LANGCHAIN_SESSION", "default")
)
# set max_workers to 1 to process tasks in order # set max_workers to 1 to process tasks in order
self.executor = ThreadPoolExecutor(max_workers=1) self.executor = ThreadPoolExecutor(max_workers=1)
self.client = client or LangChainPlusClient() self.client = client or LangChainPlusClient()
@ -103,7 +105,7 @@ class LangChainTracer(BaseTracer):
extra["runtime"] = get_runtime_environment() extra["runtime"] = get_runtime_environment()
run_dict["extra"] = extra run_dict["extra"] = extra
try: try:
self.client.create_run(**run_dict, session_name=self.session_name) self.client.create_run(**run_dict, project_name=self.project_name)
except Exception as e: except Exception as e:
# Errors are swallowed by the thread executor so we need to log them here # Errors are swallowed by the thread executor so we need to log them here
log_error_once("post", e) log_error_once("post", e)

64
langchain/client/runner_utils.py
View File

@ -237,18 +237,18 @@ async def _gather_with_concurrency(
return results return results
async def _tracer_initializer(session_name: Optional[str]) -> Optional[LangChainTracer]: async def _tracer_initializer(project_name: Optional[str]) -> Optional[LangChainTracer]:
""" """
Initialize a tracer to share across tasks. Initialize a tracer to share across tasks.
Args: Args:
session_name: The session name for the tracer. project_name: The project name for the tracer.
Returns: Returns:
A LangChainTracer instance with an active session. A LangChainTracer instance with an active project.
""" """
if session_name: if project_name:
tracer = LangChainTracer(session_name=session_name) tracer = LangChainTracer(project_name=project_name)
return tracer return tracer
else: else:
return None return None
@ -260,12 +260,12 @@ async def arun_on_examples(
*, *,
concurrency_level: int = 5, concurrency_level: int = 5,
num_repetitions: int = 1, num_repetitions: int = 1,
session_name: Optional[str] = None, project_name: Optional[str] = None,
verbose: bool = False, verbose: bool = False,
tags: Optional[List[str]] = None, tags: Optional[List[str]] = None,
) -> Dict[str, Any]: ) -> Dict[str, Any]:
""" """
Run the chain on examples and store traces to the specified session name. Run the chain on examples and store traces to the specified project name.
Args: Args:
examples: Examples to run the model or chain over examples: Examples to run the model or chain over
@ -276,7 +276,7 @@ async def arun_on_examples(
num_repetitions: Number of times to run the model on each example. num_repetitions: Number of times to run the model on each example.
This is useful when testing success rates or generating confidence This is useful when testing success rates or generating confidence
intervals. intervals.
session_name: Session name to use when tracing runs. project_name: Project name to use when tracing runs.
verbose: Whether to print progress. verbose: Whether to print progress.
tags: Tags to add to the traces. tags: Tags to add to the traces.
@ -307,7 +307,7 @@ async def arun_on_examples(
await _gather_with_concurrency( await _gather_with_concurrency(
concurrency_level, concurrency_level,
functools.partial(_tracer_initializer, session_name), functools.partial(_tracer_initializer, project_name),
*(functools.partial(process_example, e) for e in examples), *(functools.partial(process_example, e) for e in examples),
) )
return results return results
@ -386,11 +386,11 @@ def run_on_examples(
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
*, *,
num_repetitions: int = 1, num_repetitions: int = 1,
session_name: Optional[str] = None, project_name: Optional[str] = None,
verbose: bool = False, verbose: bool = False,
tags: Optional[List[str]] = None, tags: Optional[List[str]] = None,
) -> Dict[str, Any]: ) -> Dict[str, Any]:
"""Run the chain on examples and store traces to the specified session name. """Run the chain on examples and store traces to the specified project name.
Args: Args:
examples: Examples to run model or chain over. examples: Examples to run model or chain over.
@ -401,14 +401,14 @@ def run_on_examples(
num_repetitions: Number of times to run the model on each example. num_repetitions: Number of times to run the model on each example.
This is useful when testing success rates or generating confidence This is useful when testing success rates or generating confidence
intervals. intervals.
session_name: Session name to use when tracing runs. project_name: Project name to use when tracing runs.
verbose: Whether to print progress. verbose: Whether to print progress.
tags: Tags to add to the run traces. tags: Tags to add to the run traces.
Returns: Returns:
A dictionary mapping example ids to the model outputs. A dictionary mapping example ids to the model outputs.
""" """
results: Dict[str, Any] = {} results: Dict[str, Any] = {}
tracer = LangChainTracer(session_name=session_name) if session_name else None tracer = LangChainTracer(project_name=project_name) if project_name else None
for i, example in enumerate(examples): for i, example in enumerate(examples):
result = run_llm_or_chain( result = run_llm_or_chain(
example, example,
@ -425,13 +425,13 @@ def run_on_examples(
return results return results
def _get_session_name( def _get_project_name(
session_name: Optional[str], project_name: Optional[str],
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
dataset_name: str, dataset_name: str,
) -> str: ) -> str:
if session_name is not None: if project_name is not None:
return session_name return project_name
current_time = datetime.now().strftime("%Y-%m-%d-%H-%M-%S") current_time = datetime.now().strftime("%Y-%m-%d-%H-%M-%S")
if isinstance(llm_or_chain_factory, BaseLanguageModel): if isinstance(llm_or_chain_factory, BaseLanguageModel):
model_name = llm_or_chain_factory.__class__.__name__ model_name = llm_or_chain_factory.__class__.__name__
@ -446,13 +446,13 @@ async def arun_on_dataset(
*, *,
concurrency_level: int = 5, concurrency_level: int = 5,
num_repetitions: int = 1, num_repetitions: int = 1,
session_name: Optional[str] = None, project_name: Optional[str] = None,
verbose: bool = False, verbose: bool = False,
client: Optional[LangChainPlusClient] = None, client: Optional[LangChainPlusClient] = None,
tags: Optional[List[str]] = None, tags: Optional[List[str]] = None,
) -> Dict[str, Any]: ) -> Dict[str, Any]:
""" """
Run the chain on a dataset and store traces to the specified session name. Run the chain on a dataset and store traces to the specified project name.
Args: Args:
client: Client to use to read the dataset. client: Client to use to read the dataset.
@ -464,7 +464,7 @@ async def arun_on_dataset(
num_repetitions: Number of times to run the model on each example. num_repetitions: Number of times to run the model on each example.
This is useful when testing success rates or generating confidence This is useful when testing success rates or generating confidence
intervals. intervals.
session_name: Name of the session to store the traces in. project_name: Name of the project to store the traces in.
Defaults to {dataset_name}-{chain class name}-{datetime}. Defaults to {dataset_name}-{chain class name}-{datetime}.
verbose: Whether to print progress. verbose: Whether to print progress.
client: Client to use to read the dataset. If not provided, a new client: Client to use to read the dataset. If not provided, a new
@ -472,11 +472,10 @@ async def arun_on_dataset(
tags: Tags to add to each run in the sesssion. tags: Tags to add to each run in the sesssion.
Returns: Returns:
A dictionary containing the run's session name and the resulting model outputs. A dictionary containing the run's project name and the resulting model outputs.
""" """
client_ = client or LangChainPlusClient() client_ = client or LangChainPlusClient()
session_name = _get_session_name(session_name, llm_or_chain_factory, dataset_name) project_name = _get_project_name(project_name, llm_or_chain_factory, dataset_name)
client_.create_session(session_name, mode="eval")
dataset = client_.read_dataset(dataset_name=dataset_name) dataset = client_.read_dataset(dataset_name=dataset_name)
examples = client_.list_examples(dataset_id=str(dataset.id)) examples = client_.list_examples(dataset_id=str(dataset.id))
@ -485,12 +484,12 @@ async def arun_on_dataset(
llm_or_chain_factory, llm_or_chain_factory,
concurrency_level=concurrency_level, concurrency_level=concurrency_level,
num_repetitions=num_repetitions, num_repetitions=num_repetitions,
session_name=session_name, project_name=project_name,
verbose=verbose, verbose=verbose,
tags=tags, tags=tags,
) )
return { return {
"session_name": session_name, "project_name": project_name,
"results": results, "results": results,
} }
@ -500,12 +499,12 @@ def run_on_dataset(
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
*, *,
num_repetitions: int = 1, num_repetitions: int = 1,
session_name: Optional[str] = None, project_name: Optional[str] = None,
verbose: bool = False, verbose: bool = False,
client: Optional[LangChainPlusClient] = None, client: Optional[LangChainPlusClient] = None,
tags: Optional[List[str]] = None, tags: Optional[List[str]] = None,
) -> Dict[str, Any]: ) -> Dict[str, Any]:
"""Run the chain on a dataset and store traces to the specified session name. """Run the chain on a dataset and store traces to the specified project name.
Args: Args:
dataset_name: Name of the dataset to run the chain on. dataset_name: Name of the dataset to run the chain on.
@ -516,7 +515,7 @@ def run_on_dataset(
num_repetitions: Number of times to run the model on each example. num_repetitions: Number of times to run the model on each example.
This is useful when testing success rates or generating confidence This is useful when testing success rates or generating confidence
intervals. intervals.
session_name: Name of the session to store the traces in. project_name: Name of the project to store the traces in.
Defaults to {dataset_name}-{chain class name}-{datetime}. Defaults to {dataset_name}-{chain class name}-{datetime}.
verbose: Whether to print progress. verbose: Whether to print progress.
client: Client to use to access the dataset. If None, a new client client: Client to use to access the dataset. If None, a new client
@ -524,22 +523,21 @@ def run_on_dataset(
tags: Tags to add to each run in the sesssion. tags: Tags to add to each run in the sesssion.
Returns: Returns:
A dictionary containing the run's session name and the resulting model outputs. A dictionary containing the run's project name and the resulting model outputs.
""" """
client_ = client or LangChainPlusClient() client_ = client or LangChainPlusClient()
session_name = _get_session_name(session_name, llm_or_chain_factory, dataset_name) project_name = _get_project_name(project_name, llm_or_chain_factory, dataset_name)
client_.create_session(session_name, mode="eval")
dataset = client_.read_dataset(dataset_name=dataset_name) dataset = client_.read_dataset(dataset_name=dataset_name)
examples = client_.list_examples(dataset_id=str(dataset.id)) examples = client_.list_examples(dataset_id=str(dataset.id))
results = run_on_examples( results = run_on_examples(
examples, examples,
llm_or_chain_factory, llm_or_chain_factory,
num_repetitions=num_repetitions, num_repetitions=num_repetitions,
session_name=session_name, project_name=project_name,
verbose=verbose, verbose=verbose,
tags=tags, tags=tags,
) )
return { return {
"session_name": session_name, "project_name": project_name,
"results": results, "results": results,
} }

1328
langchain/experimental/client/tracing_datasets.ipynb
View File

File diff suppressed because it is too large Load Diff

10
poetry.lock generated
View File

@ -4362,13 +4362,13 @@ tests = ["doctest", "pytest", "pytest-mock"]
[[package]] [[package]]
name = "langchainplus-sdk" name = "langchainplus-sdk"
version = "0.0.15" version = "0.0.17"
description = "Client library to connect to the LangChainPlus LLM Tracing and Evaluation Platform." description = "Client library to connect to the LangSmith LLM Tracing and Evaluation Platform."
optional = false optional = false
python-versions = ">=3.8.1,<4.0" python-versions = ">=3.8.1,<4.0"
files = [ files = [
{file = "langchainplus_sdk-0.0.15-py3-none-any.whl", hash = "sha256:e69bdbc8af6007ef2f774248d2483bbaf2d75712b1acc9ea50eda3b9f6dc567d"}, {file = "langchainplus_sdk-0.0.17-py3-none-any.whl", hash = "sha256:899675fe850bb0829691ce7643d5c3b4425de1535b6f2d6ce1e5f5457ffb05bf"},
{file = "langchainplus_sdk-0.0.15.tar.gz", hash = "sha256:ce40e9e3b6d42741f0a2aa89f83a12f2648f38690a9dd57e5fe3a56f2f232908"}, {file = "langchainplus_sdk-0.0.17.tar.gz", hash = "sha256:6520c864a23dcadbe6fb7233a117347f6acc32725a97758e59354704c50de303"},
] ]
[package.dependencies] [package.dependencies]
@ -11771,4 +11771,4 @@ text-helpers = ["chardet"]
[metadata] [metadata]
lock-version = "2.0" lock-version = "2.0"
python-versions = ">=3.8.1,<4.0" python-versions = ">=3.8.1,<4.0"
content-hash = "09d46ad12369c6a16513558618553623cd520c2855bff3b8fe8248e1b18cbb94" content-hash = "6e495e4f58127a5d2001385404b973896e275f5ca71a6ebe856cb114977189d1"

2
pyproject.toml
View File

@ -106,7 +106,7 @@ pyspark = {version = "^3.4.0", optional = true}
clarifai = {version = "9.1.0", optional = true} clarifai = {version = "9.1.0", optional = true}
tigrisdb = {version = "^1.0.0b6", optional = true} tigrisdb = {version = "^1.0.0b6", optional = true}
nebula3-python = {version = "^3.4.0", optional = true} nebula3-python = {version = "^3.4.0", optional = true}
langchainplus-sdk = ">=0.0.13" langchainplus-sdk = ">=0.0.17"
awadb = {version = "^0.3.3", optional = true} awadb = {version = "^0.3.3", optional = true}
azure-search-documents = {version = "11.4.0a20230509004", source = "azure-sdk-dev", optional = true} azure-search-documents = {version = "11.4.0a20230509004", source = "azure-sdk-dev", optional = true}
openllm = {version = ">=0.1.6", optional = true} openllm = {version = ">=0.1.6", optional = true}

6
tests/unit_tests/client/test_runner_utils.py
View File

@ -176,7 +176,7 @@ async def test_arun_on_dataset(monkeypatch: pytest.MonkeyPatch) -> None:
{"result": f"Result for example {example.id}"} for _ in range(n_repetitions) {"result": f"Result for example {example.id}"} for _ in range(n_repetitions)
] ]
def mock_create_session(*args: Any, **kwargs: Any) -> None: def mock_create_project(*args: Any, **kwargs: Any) -> None:
pass pass
with mock.patch.object( with mock.patch.object(
@ -186,7 +186,7 @@ async def test_arun_on_dataset(monkeypatch: pytest.MonkeyPatch) -> None:
), mock.patch( ), mock.patch(
"langchain.client.runner_utils._arun_llm_or_chain", new=mock_arun_chain "langchain.client.runner_utils._arun_llm_or_chain", new=mock_arun_chain
), mock.patch.object( ), mock.patch.object(
LangChainPlusClient, "create_session", new=mock_create_session LangChainPlusClient, "create_project", new=mock_create_project
): ):
client = LangChainPlusClient(api_url="http://localhost:1984", api_key="123") client = LangChainPlusClient(api_url="http://localhost:1984", api_key="123")
chain = mock.MagicMock() chain = mock.MagicMock()
@ -195,7 +195,7 @@ async def test_arun_on_dataset(monkeypatch: pytest.MonkeyPatch) -> None:
dataset_name="test", dataset_name="test",
llm_or_chain_factory=lambda: chain, llm_or_chain_factory=lambda: chain,
concurrency_level=2, concurrency_level=2,
session_name="test_session", project_name="test_project",
num_repetitions=num_repetitions, num_repetitions=num_repetitions,
client=client, client=client,
) )