readme

C-package of "Long Short-Term Memory" for Protein classification

This is a quickstart experiment where one SCOP superfamily is trained and
tested with a withdrawn family member.
To start the test the LSTM should be compiled by running mak in the src folder.
Then a parameter file determines the architecture of the LSTM. E.g 'lsmtpars.mem12.ws9.txt'
chooses an architecture with 12 memory cells and a window size of 9 amino acids.
The experiment can now be started using e.g. the named architecture by running 

src/lstm -c lstmpars.mem12.ws9.txt

The LSTM gives out the performance on every epoch and stops after 100 epochs.
The performance metrics are Mean Square Error (mse), false negatives (fn) and
false postives (fp) in percent. The progress is also written in to the file 'out.txt'.