centersl/MinerU
1
0
Fork 0
mirror of https://github.com/opendatalab/MinerU.git synced 2026-03-27 11:08:32 +07:00
Code Issues Packages Projects Releases Wiki Activity

fix: update device handling and backend configuration in analysis scripts

Browse Source
This commit is contained in:
myhloli
2025-11-11 11:40:52 +08:00
parent 672e252506
commit 5c743dc169
2 changed files with 8 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
demo/demo.py
View File

@@ -236,5 +236,6 @@ if __name__ == '__main__':
"""To enable VLM mode, change the backend to 'vlm-xxx'"""
# parse_doc(doc_path_list, output_dir, backend="vlm-transformers") # more general.
# parse_doc(doc_path_list, output_dir, backend="vlm-mlx-engine") # faster than transformers in macOS 13.5+.
# parse_doc(doc_path_list, output_dir, backend="vlm-vllm-engine") # faster(engine).
# parse_doc(doc_path_list, output_dir, backend="vlm-vllm-engine") # faster(vllm-engine).
# parse_doc(doc_path_list, output_dir, backend="vlm-lmdeploy-engine") # faster(lmdeploy-engine).
# parse_doc(doc_path_list, output_dir, backend="vlm-http-client", server_url="http://127.0.0.1:30000") # faster(client).

16
mineru/backend/vlm/vlm_analyze.py
View File

@@ -128,24 +128,20 @@ class ModelSingleton:
if "cache_max_entry_count" not in kwargs:
kwargs["cache_max_entry_count"] = 0.5
device = kwargs.get("device", "").lower()
# 默认使用 turbomind
lm_backend = "turbomind"
device = kwargs.get("device", "cuda").lower()
# 特定设备强制使用 pytorch backend
if device in ["ascend", "maca", "camb"]:
lm_backend = "pytorch"
backend_config = PytorchEngineConfig(**kwargs)
else:
# 其他情况根据 lm_backend 参数决定,默认使用 turbomind
lm_backend = kwargs.get("lm_backend", "turbomind")
if lm_backend == "pytorch":
backend_config = PytorchEngineConfig(**kwargs)
else:
lm_backend = "turbomind" # 确保非 pytorch 时使用 turbomind
backend_config = TurbomindEngineConfig(**kwargs)
backend_config = TurbomindEngineConfig(**kwargs)
log_level = 'ERROR'
from lmdeploy.utils import get_logger
logger = get_logger('lmdeploy')
logger.setLevel(log_level)
lm_logger = get_logger('lmdeploy')
lm_logger.setLevel(log_level)
if os.getenv('TM_LOG_LEVEL') is None:
os.environ['TM_LOG_LEVEL'] = log_level