Model saving and loading method:
the saving method is the built-in xgboost, save_model() under Booster; the
loading method is the built-in xgboost, load_model() under Booster;

Error details:
‘XGBoostError: [14:03:43] C:\Users\Administrator\workspace\xgboost-win64_release_1.5.1\include\xgboost/json.h:73: Invalid cast, from Integer to Boolean’,

Reason for the error:
The error is caused by the low version of python and xgboost. Under python3.6, xgboost can only be updated to 1.5.2;

Solution:
Create a new virtual environment, specify python=3.8, and install xgboost=1.6.1.

Similar Posts:

• [Solved] 1.4pip Install xgboost Error: Command errored out with exit status
• numpy.distutils.system_info.NotFoundError: no lapack/blas resources found problem solved
• [Solved] Python virtual environment install Error: returned non zero exit status 1
• [Solved] module ‘keras.engine.topology’ has no attribute ‘load_weights_from_hdf5_group_by_name…
• Pychar can’t install the third-party library, the solution of error code non zero exit code (1)
• Using the jupyter notebook (error/usr/bin/python3: no module named ipykernel) on the local browser remote server_ launcher)
• Solution to error reporting of CONDA virtual environment due to system environment variables
• [Solved] python MATLAB Error: OSError: MATLAB Engine for Python supports Python version 2.7, 3.4, 3.5 and3.6, but your version of python is 3.8
• [Solved] Jupiter notebook failed to start Error: DLL load failed
• [Solved] jenkins Build python Program Error: ModuleNotFoundError: No module named ‘htmltestreport’