[MAP Logo]

Materials Algorithms Project
Program Library


  1. Provenance of code.
  2. Purpose of code.
  3. Specification.
  4. Description of program's operation.
  5. References.
  6. Parameter descriptions.
  7. Error indicators.
  8. Accuracy estimate.
  9. Any additional information.
  10. Example of code
  11. Auxiliary routines required.
  12. Keywords.
  13. Download source code.
  14. Links.

Provenance of Source Code

Murugananth M
Phase Transformations Group,
Department of Materials Science and Metallurgy,
University of Cambridge,
Cambridge CB2 3QZ, U.K.

Vasudevan M
Indhira Gandhi Center for Atomic Research,

The neural network program was produced by:

David MacKay,
Cavendish Laboratory,
University of Cambridge,
Madingley Road,
Cambridge, CB3 0HE, U.K.

E-mail: Murugananth

Added to MAP: February 2002.

Top | Next


To predict the ferrite number of austenitic steel welds.

Top | Next | Prev


Language: C,FORTRAN, Shell script
Product form: Source code / executables
Platform : Solaris, Linux

Complete program.

Top | Next | Prev


MAP_NEURAL_FERRITE_NUMBER contains a suite of programs which can be used to predict the ferrite number of austenitic stainless steel welds as a function of composition. The database on which this model has been built mainly consists of composition of austenitic stainless steel welds used in construction of the constitutional diagram WRC-92.

Description of the files and their usage in the present module :

Readme.txt Tells the name of the input variables and the order in which these are to be presented in a file named test.dat before using the model for predictions.
MINMAXContains the minimum and maximum ranges of each input variable.
test.datContains the input variables which will be used by accessory programs to make predictions.
data_no.cA C-file to get the number of data in test.dat file via keyboard input.
model.genA file containing the shell commands for using the module for predictions. This can be run by using the command "./model.gen" or "sh model.gen". This takes you through various stages where the data are first normalised allowing the models to generalise their predictions. Further after the predictions the results are un-normalised and written to a result file. model.gen also shows the maximum perceived noise level for the "n" models in the committee
generate44This is the executable file for the neural network program. It reads the normalised input data file, norm_test.in, and uses the weight files in subdirectory c, to find a value for ferrite number.The results are written to the temporary output file _out.
spec.c Specification file to be read by generate44.
_ot, _out, _res, _senThese files are created by generate44 and can be deleted.
c Directory where weight files for use by the generate44 pertaining to each model in the committee are stored.
outprdtDirectory where the output from generate44 are written to.
committee.datConsists of values representing the number of models in committee and number of variables used.
normtest.for Program to normalise the input data read from test.dat. Also produces the normalised input file norm_test.in. It makes use of information read in from no_of_rows.dat and committee.dat.
gencom.for This program uses the information in committee.dat and combines the predictions from the individual models, in subdirectory outprdt, to obtain an averaged value (committee prediction). The output (in normalised form) is written to com.dat.
treatout.for Program to un-normalise the committee results in com.dat and write the output predictions to unnorm_com. This file is then renamed as result.

Top | Next | Prev


  1. M. Vasudevan, M. Murugananth and A.K. Bhaduri, Mathematical Modelling of Weld Phenomena 6, eds. H. Cerjak & H.K.D.H. Bhadeshia, Inst. of Materials, London, To be published
  2. D.J.C. MacKay, 1997, Mathematical Modelling of Weld Phenomena 3, eds. H. Cerjak & H.K.D.H. Bhadeshia, Inst. of Materials, London, pp 359.
  3. D.J.C MacKay's website at http://wol.inference.phy.cam.ac.uk/mackay/README.html#Source_code

Top | Next | Prev


Input parameters

Output parameters

File named "result" with format predictions, error, predictions-error, prediction+error

Top | Next | Prev

Error Indicators


Top | Next | Prev


No information supplied.

Top | Next | Prev

Further Comments


Top | Next | Prev


1. Program text

Complete program.

2. Program data

See file test.dat

3. Program results

Refer example file  : result

Top | Next | Prev

Auxiliary Routines


Top | Next | Prev


Ferrite number, austenitic stainless steel welds, delta ferrite

Top | Next | Prev


Download source code

Top | Prev

MAP originated from a joint project of the National Physical Laboratory and the University of Cambridge.

MAP Website administration / map@msm.cam.ac.uk

Top | Program Index | MAP Homepage Valid HTML 3.2!