 myhloli
|
28b278508f
fix: add error handling for PDF conversion in common.py
|
il y a 3 semaines |
|
myhloli
|
cf59949ba9
add tiff
|
il y a 1 mois |
|
myhloli
|
c9315b8e10
Refactor suffix guessing to handle PDF extensions for AI files
|
il y a 1 mois |
|
myhloli
|
907099762f
Normalize PDF suffix handling for AI files to be case-insensitive
|
il y a 1 mois |
|
myhloli
|
2c356cccee
Fix suffix identification for AI files to correctly handle PDF extensions
|
il y a 1 mois |
|
myhloli
|
bba31191a4
fix: update backend handling to enforce correct usage of vlm engines in sync and async modes
|
il y a 2 mois |
|
myhloli
|
e60da65cca
feat: enhance file type detection using Magika for improved suffix guessing
|
il y a 2 mois |
|
myhloli
|
55eaad224d
feat: add support for vlm 2.5
|
il y a 2 mois |
|
myhloli
|
c3db578247
fix: add line sorting logic for bounding boxes in PDF processing
|
il y a 3 mois |
|
myhloli
|
9a3a314916
refactor: add support for additional keyword arguments in client and analysis functions
|
il y a 4 mois |
|
myhloli
|
67715001cf
fix: update environment variable names for formula and table enable in common.py and vlm_middle_json_mkcontent.py
|
il y a 4 mois |
|
myhloli
|
677a301a9f
fix: standardize parameter names for formula and table options across modules
|
il y a 4 mois |
|
myhloli
|
c29327fde6
feat: add support for additional image formats in file upload
|
il y a 4 mois |
|
myhloli
|
86391acf28
fix: prevent server URL assignment for non-client backends in VLM processing
|
il y a 4 mois |
|
myhloli
|
e7649e432c
refactor: modularize PDF processing and enhance async support for VLM backend
|
il y a 4 mois |
|
myhloli
|
bfaf07c69f
fix: refactor device mode and virtual VRAM size handling in client.py and common.py
|
il y a 5 mois |
|
myhloli
|
1383787bad
fix: refactor formula and table enable handling to use environment variables
|
il y a 5 mois |
|
myhloli
|
98b8c4a967
refactor: streamline formula and table enable configurations in the pipeline
|
il y a 5 mois |
|
myhloli
|
8737ebb2e2
feat: remove model path parameter from vlm_doc_analyze and streamline model loading
|
il y a 5 mois |
|
myhloli
|
4eaa85fd31
refactor: update make mode constants to improve content list handling
|
il y a 5 mois |
|
myhloli
|
02898cdd81
refactor: simplify file reading function and improve input validation
|
il y a 5 mois |
|
myhloli
|
7eed5ee9c8
refactor: streamline PDF parsing and enhance formula recognition handling
|
il y a 5 mois |
|
myhloli
|
29d262618a
refactor: add method option for PDF parsing and improve resource management
|
il y a 5 mois |
|
myhloli
|
284cec041a
refactor: replace get_file_from_repos with auto_download_and_get_model_root_path in multiple files
|
il y a 5 mois |
|
myhloli
|
a29489ef51
refactor: update config file name and enhance model path handling
|
il y a 5 mois |
|
myhloli
|
44a9cf220a
refactor: enhance model source handling and improve environment variable setup
|
il y a 5 mois |
|
myhloli
|
b0fd756625
refactor: update OCR handling and improve function parameters for clarity
|
il y a 5 mois |
|
myhloli
|
9bb257769c
refactor: reorganize imports to align with backend structure and improve clarity
|
il y a 5 mois |
|
myhloli
|
3334157f15
refactor: clean up unused OCR area calculation and update demo PDF path
|
il y a 5 mois |
|
myhloli
|
7d4ce0c380
refactor: add LLM-aided title optimization and improve config handling
|
il y a 5 mois |
