xgboostExtension
xgboostExtension copied to clipboard
bigdong89
xgboost Extension for Easy Ranking & TreeFeature
executing the example ranking code I get the following error. It seems to me a parameter is used that is not supported by xgboost. --------------------------------------------------------------------------- AttributeError Traceback (most recent call...
The fit function leads to a parameter error, maybe this has to do with different version of xgboost ``` Traceback (most recent call last): File "/home/dnachbar/python/lightgbm/xgbrank.py", line 32, in ranker.fit(X,...
change positional arguments to keyword arguments for the forward compatibility with xgboost 0.90
In sample xgbfeature_example.py, data's shape is 40×4, taget's shape is 40×1, why the index of leaf index for each tree is 40×50? I look forward to your reply. Thank you.
Conventional, the order of predict values should be consistent with the input, so I think the predict and apply int XGBRanker may has error.
Hello, The parameter "eval_set" in the function XGBRanker.fit is not passed to the xgboost train function.
https://github.com/bigdong89/xgboostExtension/blob/18d4b8bb5fc0dfb13f2b2af590fd5adc7f5232c8/xgboostextension/scorer/__init__.py#L25 On the above line, should be {} instead of None, otherwise pickling (for parallel CV) will still not work. Also note that this functionality is not present in the...
i download and run the xgbranker_example.py get an error on lines if callable(self.objective): obj = _objective_decorator(self.objective) # Use default value. Is it really not used ? xgb_options["objective"] = "rank:pairwise" i...
