Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add SummarizeImages to Aryn SDK #1189

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Add SummarizeImages to Aryn SDK #1189

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
dbc1cfd
add summarize to asdk
karanataryn Feb 17, 2025
1bb8b81
merge main
karanataryn Feb 19, 2025
4e7b101
merge main
karanataryn Feb 20, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions lib/aryn-sdk/aryn_sdk/partition/partition.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,7 @@ def partition_file(
threshold: Optional[Union[float, Literal["auto"]]] = None,
use_ocr: bool = False,
ocr_images: bool = False,
summarize_images: bool = False,
ocr_language: Optional[str] = None,
extract_table_structure: bool = False,
table_extraction_options: dict[str, Any] = {},
Expand Down Expand Up @@ -82,6 +83,7 @@ def partition_file(
default: False
ocr_images: attempt to use OCR to generate a text representation of detected images.
default: False
summarize_images: Generate a text summary of detected images using a VLM.
ocr_language: specify the language to use for OCR. If not set, the language will be english.
default: English
extract_table_structure: extract tables and their structural content.
Expand Down Expand Up @@ -157,6 +159,7 @@ def partition_file(
threshold=threshold,
use_ocr=use_ocr,
ocr_images=ocr_images,
summarize_images=summarize_images,
ocr_language=ocr_language,
extract_table_structure=extract_table_structure,
table_extraction_options=table_extraction_options,
Expand All @@ -180,6 +183,7 @@ def _partition_file_wrapper(
threshold: Optional[Union[float, Literal["auto"]]] = None,
use_ocr: bool = False,
ocr_images: bool = False,
summarize_images: bool = False,
ocr_language: Optional[str] = None,
extract_table_structure: bool = False,
table_extraction_options: dict[str, Any] = {},
Expand Down Expand Up @@ -209,6 +213,7 @@ def _partition_file_wrapper(
threshold=threshold,
use_ocr=use_ocr,
ocr_images=ocr_images,
summarize_images=summarize_images,
ocr_language=ocr_language,
extract_table_structure=extract_table_structure,
table_extraction_options=table_extraction_options,
Expand Down Expand Up @@ -236,6 +241,7 @@ def _partition_file_inner(
threshold: Optional[Union[float, Literal["auto"]]] = None,
use_ocr: bool = False,
ocr_images: bool = False,
summarize_images: bool = False,
ocr_language: Optional[str] = None,
extract_table_structure: bool = False,
table_extraction_options: dict[str, Any] = {},
Expand Down Expand Up @@ -269,6 +275,7 @@ def _partition_file_inner(
threshold=threshold,
use_ocr=use_ocr,
ocr_images=ocr_images,
summarize_images=summarize_images,
ocr_language=ocr_language,
extract_table_structure=extract_table_structure,
table_extraction_options=table_extraction_options,
Expand Down Expand Up @@ -376,6 +383,7 @@ def _json_options(
threshold: Optional[Union[float, Literal["auto"]]] = None,
use_ocr: bool = False,
ocr_images: bool = False,
summarize_images: bool = False,
ocr_language: Optional[str] = None,
extract_table_structure: bool = False,
table_extraction_options: dict[str, Any] = {},
Expand All @@ -393,6 +401,8 @@ def _json_options(
options["use_ocr"] = use_ocr
if ocr_images:
options["ocr_images"] = ocr_images
if summarize_images:
options["summarize_images"] = summarize_images
if ocr_language:
options["ocr_language"] = ocr_language
if extract_images:
Expand Down Expand Up @@ -423,6 +433,7 @@ def partition_file_async_submit(
threshold: Optional[Union[float, Literal["auto"]]] = None,
use_ocr: bool = False,
ocr_images: bool = False,
summarize_images: bool = False,
ocr_language: Optional[str] = None,
extract_table_structure: bool = False,
table_extraction_options: dict[str, Any] = {},
Expand Down Expand Up @@ -479,6 +490,7 @@ def partition_file_async_submit(
threshold=threshold,
use_ocr=use_ocr,
ocr_images=ocr_images,
summarize_images=summarize_images,
ocr_language=ocr_language,
extract_table_structure=extract_table_structure,
table_extraction_options=table_extraction_options,
Expand Down
Loading