The uploaded file is expected to be in tabular format, with each row repesenting the performance of a specific version of the model on a specfic version of the dataset.
The following pieces of information are used for each row:
  1. Column indicating the dataset version (default: "data")
  2. Column indicating the model version (default: "model")
  3. Column indicating the performance (default: "performance")
  4. (Optional) Column indicating the repetition number of the combination of model and dataset version; used to calculate confidence intervals (default: "repetition")
Example:
Data Model Performance Repetition (Optional)
1 1 0.7 1
1 1 0.8 2
1 2 0.5 1
... ... ... ...
N N 0.8 R
Input data
Data
VIGILANT Measurements download
By Repetition download