phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-31 08:20:00 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Initial implementation of RAG operator using the preprocessing endpoint.

Browse source
This commit is contained in:
ilya-kolchinsky 2025-03-05 13:43:26 +01:00
parent c2bd31eb5c
commit 16764a2f06
8 changed files with 74 additions and 37 deletions
Download patch file Download diff file Expand all files Collapse all files

14
llama_stack/apis/preprocessing/preprocessing.py
View file

@ -23,10 +23,22 @@ class PreprocessingDataType(Enum):
chunks = "chunks"
class PreprocessingDataFormat(Enum):
pdf = "pdf"
docx = "docx"
xlsx = "xlsx"
pptx = "pptx"
md = "md"
json = "json"
html = "html"
csv = "csv"
@json_schema_type
class PreprocessingInput(BaseModel):
preprocessor_input_id: str
preprocessor_input_type: Optional[PreprocessingDataType] = None
preprocessor_input_format: Optional[PreprocessingDataFormat] = None
path_or_content: str | InterleavedContent | URL
@ -52,5 +64,5 @@ class Preprocessing(Protocol):
self,
preprocessor_id: str,
preprocessor_inputs: List[PreprocessingInput],
options: PreprocessorOptions,
options: Optional[PreprocessorOptions] = None,
) -> PreprocessingResponse: ...