True, I forgot this. I made a note for myself. http://wiki.nmr-relax.com/Epd_canopy#For_Linux
2014-08-26 14:59 GMT+02:00 Edward d'Auvergne <edw...@nmr-relax.com>: > If you have a look at the 'sconstruct' script, specifically in the > relax_fit() method, you will also see that you can set the > PYTHON_INCLUDE_DIR environmental variable to where ever the > corresponding Python.h file is. > > Regards, > > Edward > > On 26 August 2014 14:58, Edward d'Auvergne <edw...@nmr-relax.com> wrote: >> Hi, >> >> The problem is shown in the line: >> >> target_functions/relax_fit.c:21:20: error: Python.h: No such file or >> directory >> >> I.e. the Python headers cannot be found by the compiler. How do you >> compile this? Do you use 'scons'? Have you compiled the module >> before? Is the Python.h file in the directory >> /sbinlab2/software/python-enthought-dis/canopy-1.4.0-full-rh5-64/Canopy_64bit/User/include/python2.7? >> >> Regards, >> >> Edward >> >> >> On 26 August 2014 14:57, Troels Emtekær Linnet <tlin...@nmr-relax.com> wrote: >>> I get this when I try to compile: >>> >>> gcc -o target_functions/c_chi2.os -c >>> -I/sbinlab2/software/python- >>> enthought-dis/canopy-1.4.0-full-rh5-64/Canopy_64bit/User/include/python2.7 >>> -fPIC target_functions/c_chi2.c >>> gcc -o target_functions/exponential.os -c >>> -I/sbinlab2/software/python-enthought-dis/canopy-1.4.0-full-rh5-64/Canopy_64bit/User/include/python2.7 >>> -fPIC target_functions/exponential.c >>> gcc -o target_functions/relax_fit.os -c >>> -I/sbinlab2/software/python-enthought-dis/canopy-1.4.0-full-rh5-64/Canopy_64bit/User/include/python2.7 >>> -fPIC target_functions/relax_fit.c >>> target_functions/relax_fit.c:21:20: error: Python.h: No such file or >>> directory >>> target_functions/relax_fit.c:31: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before '*' token >>> target_functions/relax_fit.c:80: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before '*' token >>> target_functions/relax_fit.c:118: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before '*' token >>> target_functions/relax_fit.c:168: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before '*' token >>> target_functions/relax_fit.c:179: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before '*' token >>> target_functions/relax_fit.c:196: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before '*' token >>> target_functions/relax_fit.c:244: error: expected '=', ',', ';', 'asm' >>> or '__attribute__' before 'relax_fit_methods' >>> target_functions/relax_fit.c: In function 'initrelax_fit': >>> target_functions/relax_fit.c:305: error: 'relax_fit_methods' >>> undeclared (first use in this function) >>> target_functions/relax_fit.c:305: error: (Each undeclared identifier >>> is reported only once >>> target_functions/relax_fit.c:305: error: for each function it appears in.) >>> scons: *** [target_functions/relax_fit.os] Error 1 >>> scons: building terminated because of errors. >>> >>> 2014-08-26 13:30 GMT+02:00 Edward d'Auvergne <edw...@nmr-relax.com>: >>>> The minfx equivalent algorithms as scipy.optimize.fmin_cg() and >>>> scipy.optimize.fmin_ncg(), as well as the other scipy.optimize >>>> functions are now working with the exponential curve-fitting in relax >>>> as the gradient is now implemented. None of the Scipy algorithms >>>> require the Hessian. They really need to implement the Newton >>>> optimisation technique to be taken seriously. This is quite different >>>> to NCG: >>>> >>>> http://home.gna.org/minfx/minfx.ncg-module.html >>>> http://home.gna.org/minfx/minfx.newton-module.html >>>> >>>> Regards, >>>> >>>> Edward >>>> >>>> >>>> On 26 August 2014 13:23, <tlin...@nmr-relax.com> wrote: >>>>> Author: tlinnet >>>>> Date: Tue Aug 26 13:23:47 2014 >>>>> New Revision: 25287 >>>>> >>>>> URL: http://svn.gna.org/viewcvs/relax?rev=25287&view=rev >>>>> Log: >>>>> Removed all code regarding scipy.optimize fmin_cg and fmin_ncg. >>>>> >>>>> This problem should soon be able to be solved with minfx. >>>>> >>>>> task #7822(https://gna.org/task/index.php?7822): Implement user function >>>>> to estimate R2eff and associated errors for exponential curve fitting. >>>>> >>>>> Modified: >>>>> trunk/specific_analyses/relax_disp/estimate_r2eff.py >>>>> >>>>> Modified: trunk/specific_analyses/relax_disp/estimate_r2eff.py >>>>> URL: >>>>> http://svn.gna.org/viewcvs/relax/trunk/specific_analyses/relax_disp/estimate_r2eff.py?rev=25287&r1=25286&r2=25287&view=diff >>>>> ============================================================================== >>>>> --- trunk/specific_analyses/relax_disp/estimate_r2eff.py (original) >>>>> +++ trunk/specific_analyses/relax_disp/estimate_r2eff.py Tue Aug >>>>> 26 13:23:47 2014 >>>>> @@ -45,7 +45,7 @@ >>>>> # Scipy installed. >>>>> if scipy_module: >>>>> # Import leastsq. >>>>> - from scipy.optimize import fmin_cg, fmin_ncg, leastsq >>>>> + from scipy.optimize import leastsq >>>>> >>>>> >>>>> class Exp: >>>>> @@ -424,7 +424,6 @@ >>>>> >>>>> # 'minfx' >>>>> # 'scipy.optimize.leastsq' >>>>> -# 'scipy.optimize.fmin_cg' >>>>> def estimate_r2eff(spin_id=None, ftol=1e-15, xtol=1e-15, >>>>> maxfev=10000000, factor=100.0, method='minfx', verbosity=1): >>>>> """Estimate r2eff and errors by exponential curve fitting with >>>>> scipy.optimize.leastsq. >>>>> >>>>> @@ -540,10 +539,6 @@ >>>>> # Acquire results. >>>>> results = minimise_leastsq(E=E) >>>>> >>>>> - elif method == 'scipy.optimize.fmin_cg': >>>>> - # Acquire results. >>>>> - results = minimise_fmin_cg(E=E) >>>>> - >>>>> elif method == 'minfx': >>>>> # Acquire results. >>>>> results = minimise_minfx(E=E) >>>>> @@ -715,46 +710,6 @@ >>>>> return results >>>>> >>>>> >>>>> -def minimise_fmin_cg(E=None): >>>>> - """Estimate r2eff and errors by exponential curve fitting with >>>>> scipy.optimize.fmin_cg. >>>>> - >>>>> - Unconstrained minimization of a function using the Newton-CG method. >>>>> - >>>>> - @keyword E: The Exponential function class, which contain data >>>>> and functions. >>>>> - @type E: class >>>>> - @return: Packed list with optimised parameter, estimated >>>>> parameter error, chi2, iter_count, f_count, g_count, h_count, warning >>>>> - @rtype: list >>>>> - """ >>>>> - >>>>> - # Check that scipy.optimize.leastsq is available. >>>>> - if not scipy_module: >>>>> - raise RelaxError("scipy module is not available.") >>>>> - >>>>> - # Initial guess for minimisation. Solved by linear least squares. >>>>> - x0 = E.estimate_x0_exp() >>>>> - >>>>> - # Define function to minimise. Use errors as weights in the >>>>> minimisation. >>>>> - use_weights = True >>>>> - >>>>> - if use_weights: >>>>> - func = E.func_exp_weighted_general >>>>> - dfunc = E.func_exp_weighted_grad >>>>> - d2func = E.func_exp_weighted_hess >>>>> - >>>>> - # There are no args to the function, since values and times are >>>>> stored in the class. >>>>> - args=() >>>>> - >>>>> - gfk = dfunc(x0) >>>>> - deltak = numpy.dot(gfk, gfk) >>>>> - >>>>> - # Cannot get this to work. >>>>> - >>>>> - #xopt, fopt, fcalls, gcalls, hcalls, warnflag = fmin_ncg(f=func, >>>>> x0=x0, fprime=dfunc, fhess=None, args=args, avextol=1e-05, >>>>> epsilon=1.4901161193847656e-08, maxiter=maxfev, full_output=1, disp=1, >>>>> retall=0, callback=None) >>>>> - #test = fmin_ncg(f=func, x0=x0, fprime=dfunc, fhess=d2func, >>>>> args=args, avextol=1e-05, epsilon=1.4901161193847656e-08, maxiter=maxfev) >>>>> - #fmin_cg(f, x0, fprime=None, args=(), gtol=1e-5, norm=Inf, >>>>> epsilon=_epsilon, maxiter=None, full_output=0, disp=1, retall=0, >>>>> callback=None): >>>>> - #fmin_cg(f=func, x0=x0, fprime=dfunc, args=args, gtol=1e-5) >>>>> - >>>>> - >>>>> def minimise_minfx(E=None): >>>>> """Estimate r2eff and errors by minimising with minfx. >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> relax (http://www.nmr-relax.com) >>>>> >>>>> This is the relax-commits mailing list >>>>> relax-comm...@gna.org >>>>> >>>>> To unsubscribe from this list, get a password >>>>> reminder, or change your subscription options, >>>>> visit the list information page at >>>>> https://mail.gna.org/listinfo/relax-commits >>>> >>>> _______________________________________________ >>>> relax (http://www.nmr-relax.com) >>>> >>>> This is the relax-devel mailing list >>>> relax-devel@gna.org >>>> >>>> To unsubscribe from this list, get a password >>>> reminder, or change your subscription options, >>>> visit the list information page at >>>> https://mail.gna.org/listinfo/relax-devel _______________________________________________ relax (http://www.nmr-relax.com) This is the relax-devel mailing list relax-devel@gna.org To unsubscribe from this list, get a password reminder, or change your subscription options, visit the list information page at https://mail.gna.org/listinfo/relax-devel
