MCPcopy Create free account
hub / github.com/antmachineintelligence/mtgbmcode / eval_valid

Method eval_valid

python-package/lightgbmmt/basic.py:2196–2226  ·  view source on GitHub ↗

Evaluate for validation data. Parameters ---------- feval : callable or None, optional (default=None) Customized evaluation function. Should accept two parameters: preds, valid_data, and return (eval_name, eval_result, is_higher_better) or

(self, feval=None)

Source from the content-addressed store, hash-verified

2194 return self.__inner_eval(self._train_data_name, 0, feval)
2195
2196 def eval_valid(self, feval=None):
2197 """Evaluate for validation data.
2198
2199 Parameters
2200 ----------
2201 feval : callable or None, optional (default=None)
2202 Customized evaluation function.
2203 Should accept two parameters: preds, valid_data,
2204 and return (eval_name, eval_result, is_higher_better) or list of such tuples.
2205
2206 preds : list or numpy 1-D array
2207 The predicted values.
2208 valid_data : Dataset
2209 The validation dataset.
2210 eval_name : string
2211 The name of evaluation function (without whitespaces).
2212 eval_result : float
2213 The eval result.
2214 is_higher_better : bool
2215 Is eval result higher better, e.g. AUC is ``is_higher_better``.
2216
2217 For multi-class task, the preds is group by class_id first, then group by row_id.
2218 If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
2219
2220 Returns
2221 -------
2222 result : list
2223 List with evaluation results.
2224 """
2225 return [item for i in range_(1, self.__num_dataset)
2226 for item in self.__inner_eval(self.name_valid_sets[i - 1], i, feval)]
2227
2228 def save_model(self, filename, num_iteration=None, start_iteration=0):
2229 """Save Booster to file.

Callers 3

trainFunction · 0.95
testMethod · 0.95
cvFunction · 0.80

Calls 1

__inner_evalMethod · 0.95

Tested by 1

testMethod · 0.76