Update xgboost to 2.0.3 or newer, and support Python 3.9 or later

[sapatel2]

Chaquopy version

15.0.1

Relevant parts of your build.gradle file

 python {
        pip {
            install "xgboost==2.0.3"
            install "numpy"
        }
    }

Relevant parts of your code

import pickle
import xgboost as xgb

def print_model_variables(model):
    model_vars = vars(model)
    for var_name, var_value in model_vars.items():
        print(f"{var_name}: {var_value}")

def test(input):
    with open('/data/user/0/com.chaquo.myapplication/files/pm_model.pkl','rb') as f:
         pickle_model = pickle.load(f)
         print("XGBoost version:", xgb.__version__)
         print(input)
         print_model_variables(pickle_model)
         data = pickle_model.predict(input)
         print(data)
         return data

Describe your issue

issue facing in pickle_model.predict(input)
Error:- AttributeError: 'super' object has no attribute 'get_params'.

[mhsmith]

If you'd like to try building a newer version of this package yourself, follow the instructions here. And if you're successful, please make a pull request so we can add the new version to the public repository.

If anyone else needs a new version of this package, please click the thumbs up button above, and post a comment explaining why you need it.

[sapatel2]

Hello @mhsmith
i flowed your instruction but xgboost version not update.
i check your xgboost directory but still represent older lib.

[mhsmith]

When you build the package, you can change the version by editing the meta.yaml file in its recipe.

[sapatel2]

i am using in android editor.
i'm using 15.0.1 version. in build.gradle file

[mhsmith]

If you want a newer version of xgboost, you'll need to build it yourself. Follow the instructions in this comment.

[mhsmith]

I'm now working on adding official Android support to this package (dmlc/xgboost#11942).