def to_bigdl_metric(metric, loss):
metric = metric.lower()
loss_str = (loss if isinstance(loss, six.string_types) else
loss.__class__.__name__).lower()
if metric == "accuracy" or metric == "acc":
if loss_str == "sparse_categorical_crossentropy"\
or loss_str == "sparsecategoricalcrossentropy":
return metrics.SparseCategoricalAccuracy()
elif loss_str == "categorical_crossentropy"\
or loss_str == "categoricalcrossentropy":
return metrics.CategoricalAccuracy()
elif loss_str == "binary_crossentropy"\
or loss_str == "binarycrossentropy":
return metrics.BinaryAccuracy()
else:
raise TypeError(
"Not supported combination: metric {} and loss {}".format(
metric, loss_str))
elif metric == "top5accuracy" or metric == "top5acc":
return metrics.Top5Accuracy()
elif metric == "mae":
from bigdl.optim.optimizer import MAE
return MAE()
elif metric == "auc":
return metrics.AUC()
elif metric == "loss":
return Loss()
elif metric == "treennaccuracy":
return TreeNNAccuracy()
else:
raise TypeError("Unsupported metric: %s" % metric)
def _to_bigdl_metric(metric):
metric = metric.lower()
if metric == "accuracy" or metric == "acc":
return metrics.Accuracy()
elif metric == "top5accuracy" or metric == "top5acc":
return metrics.Top5Accuracy()
elif metric == "mae":
from bigdl.optim.optimizer import MAE
return MAE()
elif metric == "auc":
return metrics.AUC()
elif metric == "treennaccuracy":
return TreeNNAccuracy()
else:
raise TypeError("Unsupported metric: %s" % metric)
def to_bigdl_metric(metric):
metric = metric.lower()
if metric == "accuracy" or metric == "acc":
return metrics.Accuracy()
elif metric == "top5accuracy" or metric == "top5acc":
return metrics.Top5Accuracy()
elif metric == "mae":
return MAE()
elif metric == "auc":
return metrics.AUC()
elif metric == "loss":
return Loss()
elif metric == "treennaccuracy":
return TreeNNAccuracy()
else:
raise TypeError("Unsupported metric: %s" % metric)
