Mercurial > pylearn
changeset 427:fa4a5fee53ce
Showing the path to an online version of linear regressor.
| author | Yoshua Bengio <bengioy@iro.umontreal.ca> |
|---|---|
| date | Tue, 22 Jul 2008 16:44:42 -0400 |
| parents | d7611a3811f2 |
| children | 2bde0bed1919 |
| files | linear_regression.py |
| diffstat | 1 files changed, 14 insertions(+), 8 deletions(-) [+] |
line wrap: on
line diff
--- a/linear_regression.py Tue Jul 22 15:20:25 2008 -0400 +++ b/linear_regression.py Tue Jul 22 16:44:42 2008 -0400 @@ -35,14 +35,6 @@ The predictor parameters are obtained analytically from the training set. - *** NOT IMPLEMENTED YET *** - Training can proceed sequentially (with multiple calls to update with - different disjoint subsets of the training sets). After each call to - update the predictor is ready to be used (and optimized for the union - of all the training sets passed to update since construction or since - the last call to forget). - *************************** - For each (input[t],output[t]) pair in a minibatch,:: output_t = b + W * input_t @@ -189,3 +181,17 @@ return ds +#TODO : an online version +class OnlineLinearRegression(OnlineLearningAlgorithm): + """ + Training can proceed sequentially (with multiple calls to update with + different disjoint subsets of the training sets). After each call to + update the predictor is ready to be used (and optimized for the union + of all the training sets passed to update since construction or since + the last call to forget). + """ + pass + + + +