The iris dataset is a standard machine learning dataset.

See e.g. the Wikipedia article for more details.

You can start the benchmark with the script:

  python benchmark.py [run]

Note that you need Scikit Learn to load the dataset.

The result will look like

  Iris data set has 4 inputs, 3 classes and 150 examples
  The data has been split up input training and validation set.
  Correct predictions on training set: 120/120
  Confusion matrix:
  [[ 40.   0.   0.]
  [  0.  40.   0.]
  [  0.   0.  40.]]
  Correct predictions on test set: 30/30
  Confusion matrix:
  [[ 10.   0.   0.]
  [  0.  10.   0.]
  [  0.   0.  10.]]