update inference scripts

[zhang-prog]

No description provided.

[paddle-bot]

Thanks for your contribution!

[Copilot]

Pull Request Overview

This PR updates multiple inference scripts to incorporate TensorRT conversion logic and YAML configuration support while revising file selection for model loading across Python and C++ modules. Key changes include:

• Adding YAML-based configuration loading and TensorRT conversion in utility.py.
• Adjustments in prediction and model loading logic in predict_rec.py and ocr_rec.cpp.
• Updating the C++ header to incorporate YAML and improve label file handling.

Reviewed Changes

Copilot reviewed 4 out of 5 changed files in this pull request and generated 1 comment.

File	Description
tools/infer/utility.py	Added YAML import, TensorRT conversion helper functions, and config update logic.
tools/infer/predict_rec.py	Revised TextRecognizer initialization to load and process YAML config.
deploy/cpp_infer/src/ocr_rec.cpp	Changed model file selection logic to prefer inference.json if available.
deploy/cpp_infer/include/ocr_rec.h	Updated header includes and modified label file reading with YAML config.

Files not reviewed (1)

• deploy/cpp_infer/CMakeLists.txt: Language not supported

Comments suppressed due to low confidence (1)

tools/infer/utility.py:293

• The variable 'file_name' is used here but not defined in this scope; consider using a defined identifier or passing it as a parameter.

trt_save_path = f"{model_dir}/.cache/trt/{file_name}"

[Copilot]

It appears the YAML file is intended to be loaded using 'yaml_file_path' (constructed as model_dir + "/inference.yml") rather than 'label_path'; switching to 'yaml_file(yaml_file_path)' should correctly open the intended config file.

Suggested change

	std::ifstream yaml_file(label_path);
	std::ifstream yaml_file(yaml_file_path);