The parameters defined in the XGboost fit function (decoders.py line 725-730) are no longer consistent with XGBoost library. 'objective': linearregression has been deprecated in favor of 'reg:squarederror', and the parameters 'silence' is no longer a valid input. Resulting code should look like this to avoid warnings:
param = {'objective': "reg:squarederror", #for linear output
'eval_metric': "logloss", #loglikelihood loss
'max_depth': self.max_depth, #this is the only parameter we have set, it's one of the way or regularizing
'eta': self.eta,
'seed': 2925, #for reproducibility
}
The parameters defined in the XGboost fit function (decoders.py line 725-730) are no longer consistent with XGBoost library. 'objective': linearregression has been deprecated in favor of 'reg:squarederror', and the parameters 'silence' is no longer a valid input. Resulting code should look like this to avoid warnings:
param = {'objective': "reg:squarederror", #for linear output 'eval_metric': "logloss", #loglikelihood loss 'max_depth': self.max_depth, #this is the only parameter we have set, it's one of the way or regularizing 'eta': self.eta, 'seed': 2925, #for reproducibility }
Thank you!