[MAGEEC] [BEEBS] PCA characterisation of benchmarks & evaluation of features
cssjh at bristol.ac.uk
Mon Jul 28 12:06:32 BST 2014
Indeed. For the 'filing', you could look and see if the mysql database that
Ashley set up last year is appropriate, or if we need to explore a new
approach to archiving data (this is similar to the discussion you raised
two meetings ago).
On 27 July 2014 23:24, Kerstin Eder <Kerstin.Eder at bristol.ac.uk> wrote:
> Hi Greg,
> Well done, this sounds good. I think nobody has done this for the current
> version of BEEBs. Munaaf and Craig used parts of BEEBs for their MSc
> projects, but I think they did not check in "interim" results such as the
> exctracted feature vectors. :(
> It might be worth "filing" the data you obtain from the feature extraction
> for future machine learning exercises and also "for the record". :)
> Best wishes,
> On 27/07/2014 17:42, Greg Cawthorne wrote:
> Hello team!
> I have been working on doing some Principle Component Analysis of the
> features present in the current feature set. I have created a program which
> is ready to be used to evaluate beebs!
> Now I need a comprehensive set of feature vectors and which program &
> function they came from (as we're doing it on a function by function
> basis). Once I have this data I can perform PCA and then see which features
> are the most representative of the programs and also look to see if
> different types of benchmark have distinctly different important features.
> I believe this requires running the feature extractor across all the
> current benchmarks. If anyone can or has already done this and can send me
> the data it shall be appreciated. Otherwise I shall have a look into it
> Regards, Greg.
> mageec mailing listmageec at mageec.orghttp://mageec.org/cgi-bin/mailman/listinfo/mageec
> mageec mailing list
> mageec at mageec.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the mageec