9011286 - no notes
Download it from the given repo on the ctakes site (ctakes-ytex branch) Follow instructions on the site for UMLS and all that fun stuff navigate to ctakes-distribution and extract the bin tarball Use that directory as your ctakes_home environment variable
Open up a command prompt or a mysql command line client and punch in “set global max_allowed_packet=102410241024;” and then leave that command line up when using ctakes
Create a file like this:
sle-cui -1 and d.analysis_batch = '102014-3' group by f.noteid, code ]]>cd %CTAKES_HOME% bin\setenv.bat java -cp %CLASSPATH% -Dlog4j.configuration=file:%CTAKES_HOME%\config\log4j.xml -Xmx2g org.apache.ctakes.ytex.kernel.SparseDataExporterImpl -prop %CTAKES_HOME%\data_and_exports\new_data_allnotes.xml -type sparsematrix
sparsematrix can be replaced with “weka” to get a .arff file
Sparsematrix Type: Produces 3 files: attributes.txt List of all the attributes in the sparse matrix Added subject_id to this list instance.txt A list of instanceid’s along with the patient diagnosis data.txt A list with 3 columns: Instance_id produced by ctakes (starts at 1) Sparse Matrix Entry being specified (CUI) Amount of CUI confirmed (also includes noteid’s as first value)
Paste this in a file called ".theanorc.txt" ('.txt' optional) at the C:/ level
[blas]
ldflags = -LC:\Users\Clayton\AppData\Local\Enthought\Canopy\App\appdata\canopy-1.5.2.2785.win-x86_64\Scripts -lmk2_core -lmk2_intel_thread -lmk2_rt
[gcc]
cxxflags = -IC:\MinGW\include