Blade He
|
f166e73362
|
optimize data extraction algorithm: if can't find cost numeric value from PDF page text, then extract data by Vision ChatGPT
|
2024-10-15 15:57:54 -05:00 |
|
Blade He
|
0f14bf4a7a
|
1. get document/ provider mapping data
2. optimize metrics algorithm
3. Expand max token length since switch ChatGPT4o to 2024-08-06 version.
|
2024-09-23 17:21:02 -05:00 |
|
Blade He
|
932870f406
|
support split text for this case: outputs over 4K tokens.
|
2024-09-16 12:03:13 -05:00 |
|
Blade He
|
878383a72c
|
support extract the continuous page(s) for not missing next page data which without table header.
|
2024-09-06 16:29:35 -05:00 |
|
Blade He
|
843f588015
|
support chat with image by ChatGPT4o
|
2024-08-26 11:19:07 -05:00 |
|
Blade He
|
fa46b45ad5
|
support output tables as markdown format from pdf documents
|
2024-08-19 15:49:45 -05:00 |
|
Blade He
|
424c30853c
|
initial
|
2024-08-19 09:52:13 -05:00 |