 icecraft
|
b50f742fd9
feat: add parallel evalution
|
hai 8 meses |
|
icecraft
|
3a2f86a1b6
feat: add parallel evalution
|
hai 8 meses |
 myhloli
|
6116488d88
perf(inference): optimize batch processing for different GPU memory sizes
|
hai 8 meses |
|
myhloli
|
0b05dff74f
perf(inference): adjust batch ratio for high GPU memory
|
hai 8 meses |
|
myhloli
|
58b6ad8ca2
perf(inference): adjust batch ratio for GPU memory sizes
|
hai 8 meses |
|
myhloli
|
0d3304d7c9
perf(inference): adjust batch ratio for GPU memory sizes
|
hai 8 meses |
|
myhloli
|
6753df8dd3
perf(model): optimize batch analyze process
|
hai 8 meses |
|
myhloli
|
90a27ecd26
fix(magic_pdf): correct negative indexing for `end_page_id`
|
hai 8 meses |
|
Xiaomeng Zhao
|
886d253cf3
Merge pull request #1748 from nadahlberg/master
|
hai 8 meses |
 Nathan Dahlberg
|
37f3e200da
fix doc_analyze first page only
|
hai 9 meses |
|
myhloli
|
f5112e2157
fix(model): move environment variable settings to global scope
|
hai 9 meses |
|
myhloli
|
4021abeb98
refactor(magic_pdf): improve code structure and memory safety
|
hai 9 meses |
|
myhloli
|
7c76d3610f
refactor(model): integrate Ascend plugin for NPU support
|
hai 9 meses |
|
myhloli
|
b1ac7afdaa
perf(model): optimize batch ratio for different GPU memory sizes
|
hai 9 meses |
|
myhloli
|
29e7a9488f
perf(model): adjust batch ratio for different GPU memory sizes
|
hai 9 meses |
|
myhloli
|
d1af456621
perf(model): adjust batch ratio for GPU memory range
|
hai 9 meses |
|
myhloli
|
b6710b9988
fix(magic_pdf): correct batch ratio conditions for GPU memory
|
hai 10 meses |
|
myhloli
|
55447c8b64
perf(magic_pdf): optimize batch processing for GPU
|
hai 10 meses |
|
myhloli
|
037736fbaf
perf(magic_pdf): adjust batch ratio calculation for GPU memory
|
hai 10 meses |
|
myhloli
|
e74a2960fc
refactor(magic_pdf): adjust VRAM allocation and MFR batch size- Update VRAM allocation logic to use 'VIRTUAL_VRAM_SIZE' environment variable
|
hai 10 meses |
|
myhloli
|
052a4d72ed
perf(magic_pdf): optimize batch ratio calculation for GPU
|
hai 10 meses |
|
myhloli
|
49d140c55b
perf(model): adjust batch size for layout and formula detection
|
hai 10 meses |
|
myhloli
|
db8be9745d
fix(magic_pdf): limit batch ratio for GPU memory
|
hai 10 meses |
|
myhloli
|
f209ddea62
fix(magic_pdf): correct end page index and improve error handling
|
hai 10 meses |
|
myhloli
|
f350222614
feat(model): improve batch analysis logic and support npu
|
hai 10 meses |
|
myhloli
|
50f4841716
refactor(device): optimize memory cleaning and device selection
|
hai 10 meses |
|
icecraft
|
b2887ca0aa
refactor: refactor code
|
hai 11 meses |
|
myhloli
|
20438bd2b7
feat(language-detection): add YOLOv11 language detection model
|
hai 11 meses |
|
myhloli
|
9e4ebea939
refactor(magic_pdf): remove YOLO_VERBOSE setting and update YOLOv8 prediction verbosity
|
hai 11 meses |
|
myhloli
|
dd7f6781f6
fix(magic_pdf): disable PaddlePaddle signal handler
|
hai 11 meses |
