If i have understood your question correctly, you can use PMML

Open Neural Network Exchange (ONNX)

To use a simple Python-trained machine learning model in C++, you can follow these steps:

1. Export the trained model: In Python, you can use a library like joblib or pickle to save your trained model as a file. These libraries provide functions like `joblib.dump()` or `pickle.dump()` to serialize your model and save it to disk.

2. Load the model in C++: In your C++ code, you'll need to load the saved model. C++ doesn't have built-in support for loading Python models, so you'll need to use a library that provides this functionality. One popular library is TensorFlow's C++ API, which allows you to load and run trained models. Another option is to use a library like XGBoost or LightGBM, which have C++ APIs for loading and using their respective models.

3. Prepare input data: Once you have the model loaded in C++, you'll need to prepare the input data in a format that the model expects. This may involve converting the data to the appropriate data types and applying any necessary preprocessing steps.

4. Make predictions: Use the loaded model to make predictions on the input data. The specific method for making predictions will depend on the library and model you're using. Consult the documentation of the library or model to find the appropriate functions or methods to use.

5. Process the output: Once you have the predictions, you can process them in your C++ code as needed. This might involve further calculations, displaying the results, or using the predictions for other purposes.

It's worth noting that integrating Python and C++ code can have some challenges, such as differences in data types and memory management. You may need to pay attention to these aspects and ensure proper data handling between the two languages.

Overall, using a trained Python machine learning model in C++ requires exporting the model from Python, loading it in C++ using a compatible library, preparing the input data, making predictions, and processing the results in your C++ code.