Skip to content

πŸ“„ Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.

fix(rapidocr_onnxruntime): Optimize prompt when using DirctML

SWHL91e40ad4c337
diff --git a/python/rapidocr_onnxruntime/utils.py b/python/rapidocr_onnxruntime/utils.py
index 7030d7f..a7958c3 100644
--- a/python/rapidocr_onnxruntime/utils.py
+++ b/python/rapidocr_onnxruntime/utils.py
@@ -2,7 +2,6 @@
# @Author: SWHL
# @Contact: [email protected]
import argparse
-import importlib
import math
import os
import platform
@@ -79,18 +78,17 @@ class OrtInferSession:
"cudnn_conv_algo_search": "EXHAUSTIVE",
"do_copy_in_default_stream": True,
}
+ if self.use_cuda:
+ EP_list.insert(0, (CUDA_EP, cuda_provider_opts))
# check windows 10 or above
self.use_directml = (
- platform.system() == "Windows"
- and platform.release().split(".")[0] >= "10"
+ self.cfg_use_dml
+ and platform.system() == "Windows"
+ and int(platform.release().split(".")[0]) >= 10
and DIRECTML_EP in had_providers
- and self.cfg_use_dml
)
- if self.use_cuda:
- EP_list.insert(0, (CUDA_EP, cuda_provider_opts))
- elif self.use_directml:
- self._verfiy_dml()
+ if self.use_directml:
print(
"Windows 10 or above detected, try to use DirectML as primary provider"
)
@@ -100,14 +98,6 @@ class OrtInferSession:
EP_list.insert(0, (DIRECTML_EP, directml_options))
return EP_list
- def _verfiy_dml(self):
- try:
- importlib.import_module("onnxruntime-directml")
- except ModuleNotFoundError as exc:
- raise ModuleNotFoundError(
- "If there are other onnxruntime packages installed, please use pip uninstall onnxruntime to uninstall them first. \nThen install the package using DirectML through pip install onnxruntime-directml."
- ) from exc
-
def _verify_providers(self) -> None:
session_providers = self.session.get_providers()
@@ -318,7 +308,12 @@ def init_args():
det_group = parser.add_argument_group(title="Det")
det_group.add_argument("--det_use_cuda", action="store_true", default=False)
- det_group.add_argument("--det_use_dml", action="store_true", default=False)
+ det_group.add_argument(
+ "--det_use_dml",
+ action="store_true",
+ default=False,
+ help="Whether to use DirectML. The prerequisite is: the operating system is Windows 10+. First uninstall all onnxruntime packages and install only the onnxruntime-directml library.",
+ )
det_group.add_argument("--det_model_path", type=str, default=None)
det_group.add_argument("--det_limit_side_len", type=float, default=736)
det_group.add_argument(
@@ -336,7 +331,12 @@ def init_args():
cls_group = parser.add_argument_group(title="Cls")
cls_group.add_argument("--cls_use_cuda", action="store_true", default=False)
- cls_group.add_argument("--cls_use_dml", action="store_true", default=False)
+ cls_group.add_argument(
+ "--cls_use_dml",
+ action="store_true",
+ default=False,
+ help="Whether to use DirectML. The prerequisite is: the operating system is Windows 10+. First uninstall all onnxruntime packages and install only the onnxruntime-directml library.",
+ )
cls_group.add_argument("--cls_model_path", type=str, default=None)
cls_group.add_argument("--cls_image_shape", type=list, default=[3, 48, 192])
cls_group.add_argument("--cls_label_list", type=list, default=["0", "180"])
@@ -345,7 +345,12 @@ def init_args():
rec_group = parser.add_argument_group(title="Rec")
rec_group.add_argument("--rec_use_cuda", action="store_true", default=False)
- rec_group.add_argument("--rec_use_dml", action="store_true", default=False)
+ rec_group.add_argument(
+ "--rec_use_dml",
+ action="store_true",
+ default=False,
+ help="Whether to use DirectML. The prerequisite is: the operating system is Windows 10+. First uninstall all onnxruntime packages and install only the onnxruntime-directml library.",
+ )
rec_group.add_argument("--rec_model_path", type=str, default=None)
rec_group.add_argument("--rec_keys_path", type=str, default=None)
rec_group.add_argument("--rec_img_shape", type=list, default=[3, 48, 320])