phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-15 17:44:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix: Call pandas.read_* in a seperate thread (#1698)

Browse source 
These block on io reads which in turn block the
server. Move them to their own thread.

Closes: #1697

# What does this PR do?
To avoid blocking the main eventloop, updates datasetio/localfs to load
data in a seperate thread

Signed-off-by: Derek Higgins <derekh@redhat.com>
This commit is contained in:
Derek Higgins 2025-03-19 17:46:37 +00:00 • committed by GitHub
parent 65ca85ba6b
commit 6949bd1999
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 11 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

8
llama_stack/providers/inline/datasetio/localfs/datasetio.py
View file

@ -35,12 +35,12 @@ class PandasDataframeDataset:
else:
return self.df.iloc[idx].to_dict()
def load(self) -> None:
async def load(self) -> None:
if self.df is not None:
return
if self.dataset_def.source.type == "uri":
self.df = get_dataframe_from_uri(self.dataset_def.source.uri)
self.df = await get_dataframe_from_uri(self.dataset_def.source.uri)
elif self.dataset_def.source.type == "rows":
self.df = pandas.DataFrame(self.dataset_def.source.rows)
else:
@ -95,7 +95,7 @@ class LocalFSDatasetIOImpl(DatasetIO, DatasetsProtocolPrivate):
) -> IterrowsResponse:
dataset_def = self.dataset_infos[dataset_id]
dataset_impl = PandasDataframeDataset(dataset_def)
dataset_impl.load()
await dataset_impl.load()
start_index = start_index or 0
@ -114,7 +114,7 @@ class LocalFSDatasetIOImpl(DatasetIO, DatasetsProtocolPrivate):
async def append_rows(self, dataset_id: str, rows: List[Dict[str, Any]]) -> None:
dataset_def = self.dataset_infos[dataset_id]
dataset_impl = PandasDataframeDataset(dataset_def)
dataset_impl.load()
await dataset_impl.load()
new_rows_df = pandas.DataFrame(rows)
dataset_impl.df = pandas.concat([dataset_impl.df, new_rows_df], ignore_index=True)