develancer / empi Goto Github PK

Hi!

I have noticed the following aspect in the code and was wondering if that was intended.
The minimum and maximum scales are set thus:
double scaleMin = std::max(this->scaleMin, MIN_SCALE_IN_SAMPLES / freqSampling);
double scaleMax = std::min(this->scaleMax, sampleCount / freqSampling);
And then, the number of scales:
double root = std::sqrt(-2.0/M_PI * std::log(1.0-energyError));
double aDenomSqrt = 1.0 - energyError;
double aNominPart = energyError*(2.0-energyError)*(energyError*energyError-2*energyError+2);
double a = (1.0 + std::sqrt(aNominPart)) / (aDenomSqrt * aDenomSqrt);
double dl = log(a);

const double tMax = (sampleCount-1) / freqSampling;
const double lMin = log(scaleMin);
const double lMax = log(scaleMax);
long lCount = std::lrint((lMax - lMin)/dl + 0.5) + 1;

So it is basically just set by a. So decreasing scaleMax or increasing scaleMin, doesn't decrease the number of scales, but only increases their density. Which I would think is not necessary, since the density should be based on the error value.

What do you think?

Hello Dr. Różański,

I'm trying to install empi on MacOS, and have a quick question regarding the make empi (I'm new to c++, so this might be an easy question). I've installed cmake (3.27.7) and fftw (3.3.10_1), but when I ran make empi, I got an error showing that:
.../empi-1.0.1/src/SpectrogramCalculatorFFTW.cpp:7:10: fatal error: 'fftw3.h' file not found

I checked on the Terminal showing that the fftw3.h is linked under the folder /usr/local/include/. I was wondering if I should specify the location of this header file to c++ compiler in the build.make under empi-cpu.dir folder, or if you have any suggestions on this issue. Thank you so much.

hi, I'm trying to compile empi from source, but there are some problem.

After runing "sudo make install“, I was trying to run "sudo make install", but it prompts me “make: *** No rule to make target 'install'. Stop.”

Like is: