fix: update PaddleOCR 3.x API - replace deprecated gpu_mem parameter with device parameter

PaddleOCR 3.x changed the API:
- Removed: use_gpu=True/False and gpu_mem=<value>
- Added: device="gpu:0" or device="cpu"

Changes:
- Updated get_ocr_engine() to use device parameter
- Updated get_structure_engine() to use device parameter
- GPU mode: device="gpu:{gpu_device_id}"
- CPU mode: device="cpu"

This fixes the "ValueError: Unknown argument: gpu_mem" runtime error.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
egg
2025-11-14 09:22:56 +08:00
parent 7536f43513
commit d80d60f14b

View File

@@ -170,11 +170,12 @@ class OCRService:
logger.info(f"Initializing PaddleOCR engine for language: {lang} (GPU: {self.use_gpu})") logger.info(f"Initializing PaddleOCR engine for language: {lang} (GPU: {self.use_gpu})")
try: try:
# PaddleOCR 3.x uses 'device' parameter instead of 'use_gpu' and 'gpu_mem'
device = f"gpu:{settings.gpu_device_id}" if self.use_gpu else "cpu"
self.ocr_engines[lang] = PaddleOCR( self.ocr_engines[lang] = PaddleOCR(
use_angle_cls=True, use_angle_cls=True,
lang=lang, lang=lang,
use_gpu=self.use_gpu, device=device,
gpu_mem=int(settings.gpu_memory_fraction * 1000) if self.use_gpu else 500,
) )
logger.info(f"PaddleOCR engine ready for {lang} ({'GPU' if self.use_gpu else 'CPU'} mode)") logger.info(f"PaddleOCR engine ready for {lang} ({'GPU' if self.use_gpu else 'CPU'} mode)")
@@ -186,7 +187,7 @@ class OCRService:
self.ocr_engines[lang] = PaddleOCR( self.ocr_engines[lang] = PaddleOCR(
use_angle_cls=True, use_angle_cls=True,
lang=lang, lang=lang,
use_gpu=False, device="cpu",
) )
logger.info(f"PaddleOCR engine ready for {lang} (CPU mode - fallback)") logger.info(f"PaddleOCR engine ready for {lang} (CPU mode - fallback)")
else: else:
@@ -205,6 +206,8 @@ class OCRService:
logger.info(f"Initializing PP-StructureV3 engine (GPU: {self.use_gpu})") logger.info(f"Initializing PP-StructureV3 engine (GPU: {self.use_gpu})")
try: try:
# PaddleOCR 3.x uses 'device' parameter instead of 'use_gpu' and 'gpu_mem'
device = f"gpu:{settings.gpu_device_id}" if self.use_gpu else "cpu"
self.structure_engine = PPStructureV3( self.structure_engine = PPStructureV3(
use_doc_orientation_classify=False, use_doc_orientation_classify=False,
use_doc_unwarping=False, use_doc_unwarping=False,
@@ -212,8 +215,7 @@ class OCRService:
use_table_recognition=True, use_table_recognition=True,
use_formula_recognition=True, use_formula_recognition=True,
layout_threshold=0.5, layout_threshold=0.5,
use_gpu=self.use_gpu, device=device,
gpu_mem=int(settings.gpu_memory_fraction * 1000) if self.use_gpu else 500,
) )
logger.info(f"PP-StructureV3 engine ready ({'GPU' if self.use_gpu else 'CPU'} mode)") logger.info(f"PP-StructureV3 engine ready ({'GPU' if self.use_gpu else 'CPU'} mode)")
@@ -228,7 +230,7 @@ class OCRService:
use_table_recognition=True, use_table_recognition=True,
use_formula_recognition=True, use_formula_recognition=True,
layout_threshold=0.5, layout_threshold=0.5,
use_gpu=False, device="cpu",
) )
logger.info("PP-StructureV3 engine ready (CPU mode - fallback)") logger.info("PP-StructureV3 engine ready (CPU mode - fallback)")
else: else: