Skip to content
Permalink
Browse files

EXC: update

  • Loading branch information...
nno committed Jul 22, 2019
1 parent a1bd41a commit 8ab3d02db17103474d084f88d0391d44e8fc382f
Showing with 4 additions and 4 deletions.
  1. +1 −1 doc/source/ex_classify_lda.rst
  2. +3 −3 doc/source/nmsm2019_intro.rst
@@ -53,7 +53,7 @@ A single classification step can be visualized as follows (more advanced cross-v

.. figure:: _static/single_classification.png

*Illustration of (single-fold) classification*. A dataset (left) is split in a train dataset (top dataset) and a test set (bottom dataset), which must have no chunks in common. For training, a classifier (indicated by *f*) takes ``.samples`` and ``.sa.targets`` from the train dataset (horizontal arrow into *f*) and predicts, for the ``.samples`` in the test set (horizontal arrow into *f*), the targets of the test set (U-turn arrow). Classification accuracy can be assessed by computing how many samples in the test set were predicted correctly.
*Illustration of (single-fold) classification*. A dataset (left) is split in a train dataset (top dataset) and a test set (bottom dataset), which must have no chunks in common. For training, a classifier (indicated by *f*) takes ``.samples`` and ``.sa.targets`` from the train dataset (horizontal arrow into *f*) and predicts, for the ``.samples`` in the test set (vertical arrow into *f*), the targets of the test set (U-turn arrow). Classification accuracy can be assessed by computing how many samples in the test set were predicted correctly.

Single subject, single fold split-half classification
+++++++++++++++++++++++++++++++++++++++++++++++++++++
@@ -148,16 +148,16 @@ Monday
-------------- ---------------------------------------------------------------------------------------------------
12:30 Lunch break
-------------- ---------------------------------------------------------------------------------------------------
14:00 Split-half correlations. :doc:`ex_splithalf_correlations`
14:00 :doc:`ex_dataset_basics`
-------------- ---------------------------------------------------------------------------------------------------
15:30 Coffee break
-------------- ---------------------------------------------------------------------------------------------------
16:00-17:00 Classification analysis. :doc:`ex_classify_lda`. Optional :doc:`ex_classify_double_dipping`
16:00-17:30 Split-half correlations. :doc:`ex_splithalf_correlations`
-------------- ---------------------------------------------------------------------------------------------------
17:40-18:30 Optional: discuss your data models
-------------- ---------------------------------------------------------------------------------------------------
Tuesday
09:00 Classification with cross-validation. :doc:`ex_nfold_crossvalidation`
09:00 :doc:`ex_classify_lda`, :doc:`ex_nfold_crossvalidation`. Optional :doc:`ex_classify_double_dipping`
-------------- ---------------------------------------------------------------------------------------------------
10:30 Coffee break
-------------- ---------------------------------------------------------------------------------------------------

0 comments on commit 8ab3d02

Please sign in to comment.
You can’t perform that action at this time.