Here is a regression example: tools/demos/regression/uci-isolet/isolet.conf

To train it, you have to:

  1. Download and extract the UCI-Isolet dataset from here: datasets.tgz
  2. Modify the 'root' variable in your isolet.conf to point to your UCI-Isolet directory.
  3. Call train isolet.conf

The dataset is a spoken letter recognition dataset with 6238 training sample. It contains 26 categories and its inputs have 617 components.
While this dataset should be trained in classification mode, we use it here just to demonstrate how to train for regression.
Hence the obtained results are poor: the 'test_errors' should converge to approximately 62% error after 50 iterations.
To train on this dataset in classification mode, simply uncomment the last section of isolet.conf ('classification' section) and call train isolet.conf.

regression.txt · Last modified: 2011/12/25 09:50 (external edit)