Ask or search…
K
Comment on page

Metrics

The Tensorleap platform allows you to add metrics to be computed in each model evaluation or training cycle.
These Metrics are divided into two types:
  • Default metrics
    • MeanSquaredError
    • MeanSquaredLogarithmicError
    • MeanAbsoluteError
    • MeanAbsolutePercentageError
    • Accuracy
    • BinaryAccuracy
    • MeanIOU
  • Custom metrics
Adding a metric to the network map is done by:
  1. 1.
    Right clicking on the network map
  2. 2.
    Selecting the metric node
  3. 3.
    Selecting the type of the metric from the Node Details menu
  4. 4.
    connecting the metric block to the relevant nodes in the network map
In order to add a Custom metric it must be added to the dataset script and parsed first as described in the Custom metric section