Commit Graph

1765 Commits

Author SHA1 Message Date
Gilles Louppe 562b60ce8a EXAMPLE: merge plot_adaboost_iris into plot_forest_iris 2013-01-08 22:19:02 -08:00
Gilles Louppe 81aaedeae6 EXAMPLE: reduce the number of samples 2013-01-08 22:19:01 -08:00
Noel Dawe aa5d3d9bc0 FIX: overlapping y-axis labels 2013-01-08 22:19:01 -08:00
Noel Dawe 491379d994 FIX: import not used 2013-01-08 22:19:01 -08:00
Gilles Louppe 78c484d787 EXAMPLE: simplify n_estimators and n_samples 2013-01-08 22:19:01 -08:00
Noel Dawe b454d56bd3 implement AdaBoost 2013-01-08 16:50:38 -08:00
Alexandre Gramfort 60be687756 DOC : add doctring in plot_lasso_and_elasticnet.py 2013-01-07 11:19:50 +01:00
Andreas Mueller 685c6529b6 ENH rename proximity to dissimilarity 2013-01-07 09:41:48 +01:00
andy 1aacfbede8 ENH updated to more examples. 2013-01-07 09:41:48 +01:00
andy 791a2241aa FIX another typo, fix examples 2013-01-07 09:41:48 +01:00
Alexandre Gramfort 7b53c9a940 Merge pull request #1505 from NelleV/mds
Fixes on the MDS
2013-01-03 02:03:37 -08:00
Olivier Grisel 0d47916b42 ENH: add the squared hinge loss to the SGD loss example 2013-01-02 22:56:32 +01:00
Mathieu Blondel 2ef136c6eb Add perceptron loss to plot. 2013-01-03 03:21:46 +09:00
Nelle Varoquaux 9301f149a7 FIX the mds example did not plot the NMDS 2013-01-01 15:04:06 +01:00
Andreas Mueller ff3ea9570d DOC classifier comparison: plot data without decision boundary first, better (imho) color scheme. 2012-12-26 21:13:58 +01:00
Andreas Mueller 38b455d541 Cosmit pep8 2012-12-25 13:16:05 +01:00
Andreas Mueller 54385c7565 COSMIT featuers -> features typo 2012-12-22 16:48:17 +01:00
Olivier Grisel dbc4689625 Typo / fix in JL lemma example 2012-12-22 14:31:56 +01:00
Olivier Grisel 81a8146cdb eps=1.0 is no longer a valid value 2012-12-21 18:13:55 +01:00
Arnaud Joly 616e2b5d64 ENH Rename Bernoulli random projection to sparse random projection 2012-12-21 14:48:01 +01:00
Arnaud Joly 12c8d3be79 ENH Rename Bernoulli random projection to sparse random projection 2012-12-21 14:48:01 +01:00
Arnaud Joly e511945b80 Make the JL / RP example use the digits dataset by default 2012-12-21 14:48:00 +01:00
Arnaud Joly ccea7f1fc8 DOC clarification 2012-12-21 14:47:59 +01:00
Arnaud Joly d1a223be40 DOC fix typo in filename 2012-12-21 14:47:59 +01:00
Arnaud Joly eb67e016eb FIX typo mistakes 2012-12-21 14:47:58 +01:00
Arnaud Joly d5ed2ff94c DOC removed unrelevant paragraph(s) 2012-12-21 14:47:56 +01:00
Arnaud Joly 6cc099771c FIX example import (name changed) 2012-12-21 14:34:23 +01:00
Olivier Grisel 1e528c4df7 use the 20 newsgroups as example dataset instead 2012-12-21 14:34:21 +01:00
Olivier Grisel a850892b21 OPTIM: call murmurhash once + update test & example 2012-12-21 14:34:20 +01:00
Olivier Grisel 82cc48ffa3 working on implicit random projections using a hashing function 2012-12-21 14:34:20 +01:00
Olivier Grisel 6649d5da61 JL Example improvements 2012-12-21 14:34:20 +01:00
Olivier Grisel f13cbf4590 WIP: tweaking JL function names 2012-12-21 14:34:19 +01:00
Olivier Grisel 8840fb9b1e More work on the JL example to plot the distribution of the distortion 2012-12-21 14:34:19 +01:00
Olivier Grisel 4ca8e34173 started work on plotting the JL bounds to be used in the narrative documentation 2012-12-21 14:34:19 +01:00
Olivier Grisel 9701ce7fac remove the random projection option from the 20 newsgroups example
- the output will be dense and the models expect sparse input.
- the hyperparameters would need to be re-tuned for the new
  dimensionality. f1 scores without re-tuning is misleading

Better have dedicated examples for random projections.
2012-12-21 14:34:19 +01:00
Olivier Grisel 8eb855446a make it possible to use random projection on the 20 newsgroups classification example 2012-12-21 14:34:18 +01:00
Olivier Grisel 110842ab64 use random projections in the digits manifold example 2012-12-21 14:34:18 +01:00
Olivier Grisel e4980bc0b6 Merge pull request #1470 from kuantkid/fix_spectral_cluster_test
FIX spectral clustering test error in OSX and ADD spectral embedding to sphere examples
2012-12-19 04:57:31 -08:00
Andreas Mueller 22b52eb124 COSMIT renaming chunk_size to batch_size in MiniBatchDictionaryLearning and MiniBatchSparsePCA 2012-12-16 15:45:34 +01:00
kuantkid 0e77bcecd9 FIX spectral_embedding test erros, ADD spectral embedding to sphere examples 2012-12-13 13:05:28 +08:00
andy 5bd197cb0d DOC added note in example that digits data is to small. 2012-12-09 15:37:00 +01:00
Andreas Mueller d4eb8196e9 ENH renamed class NystromKernelApproximation to Nystrom (it is in the kernel_approximation module). Also improvements to example docstring 2012-12-09 14:14:09 +01:00
Andreas Mueller 25b60c1681 ENH Nystroem kernel approxmation 2012-12-09 14:14:09 +01:00
andy 61e2a71400 ENH fix random seed in manifold example 2012-12-08 18:43:16 +01:00
andy 0bdf3d918d COSMIT RST in manifold sphere example. 2012-12-08 18:24:32 +01:00
andy c64a6b365c FIX manifold example - sorry, my bad. 2012-12-08 18:02:25 +01:00
Andreas Mueller 1b06789bb3 COSMIT some fixes in whatsnew rst 2012-12-07 22:56:52 +00:00
Peter Prettenhofer 8897c6d206 use 100 instead of 800 n_estimators; looks the same but faster; ESLII uses 800 2012-12-03 20:37:32 +01:00
Peter Prettenhofer b7b91ef9ce Merge branch 'master' into gbrt-interactions 2012-12-02 16:05:44 +01:00
Jaques Grobler f09a579713 heading change 2012-12-02 09:36:55 +00:00