An project that trys make an approximation of an polynomial coefs using linear least square method. Right now data input is randomized inside of the program if no polynome_coefs field in config_init_data.json exists.
- a C++ compiler.
- maybe GNUplot
Right now not so much. You can check config_init_data.json for available runtime settings.
- Install vcpkg : https://vcpkg.io/en/getting-started .
Here's instructions for a Linux OS: ( for windows justbootstrap-vcpkg.bat)
git clone https://github.com/Microsoft/vcpkg.git
./vcpkg/bootstrap-vcpkg.sh
- Get sources of the project, get into the folder with project, configure and compile it. In configuration command put your path to vcpkg folder.
git clone https://github.com/Gerodote/denisov_lab1
cd denisov_lab1
cmake -S . -B ./build -DCMAKE_TOOLCHAIN_FILE=[path to vcpkg]/scripts/buildsystems/vcpkg.cmake -Dcha1_BUILD_EXECUTABLE=1 -Dcha1_ENABLE_LTO=1 -DCMAKE_BUILD_TYPE="Release" -DCMAKE_CXX_FLAGS="-march=native"
cmake --build ./build --target all --parallel 4
- Maybe you need install
gnuplotsomehow. You'll find it out when you'll try run the project exectutable:
./build/bin/Release/cha1
You should get something like this and a window with plot:
x-axis:
0 0.344828 0.689655 1.03448 1.37931 1.72414 2.06897 2.41379 2.75862 3.10345 3.44828 3.7931 4.13793 4.48276 4.82759 5.17241 5.51724 5.86207 6.2069 6.55172 6.89655 7.24138 7.58621 7.93103 8.27586 8.62069 8.96552 9.31034 9.65517 10
polynome coefs:
{ 0.375158, -0.61321, 0.257092, 0.253129 }
generated data:
0.375158 0.204655 0.157564 0.296159 0.682711 1.37949 2.44878 3.95285 5.95396 8.51439 11.6964 15.5623 20.1744 25.5948 31.886 39.1101 47.3294 56.6062 67.0028 78.5815 91.4045 105.534 121.033 137.962 156.385 176.364 197.96 221.237 246.257 273.081
vandermonde:
1 0 0 0
1 0.344828 0.118906 0.0410021
1 0.689655 0.475624 0.328017
1 1.03448 1.07015 1.10706
1 1.37931 1.9025 2.62413
1 1.72414 2.97265 5.12526
1 2.06897 4.28062 8.85645
1 2.41379 5.8264 14.0637
1 2.75862 7.60999 20.9931
1 3.10345 9.63139 29.8905
1 3.44828 11.8906 41.0021
1 3.7931 14.3876 54.5738
1 4.13793 17.1225 70.8516
1 4.48276 20.0951 90.0816
1 4.82759 23.3056 112.51
1 5.17241 26.7539 138.382
1 5.51724 30.44 167.945
1 5.86207 34.3639 201.443
1 6.2069 38.5256 239.124
1 6.55172 42.9251 281.233
1 6.89655 47.5624 328.017
1 7.24138 52.4376 379.72
1 7.58621 57.5505 436.59
1 7.93103 62.9013 498.872
1 8.27586 68.4899 566.813
1 8.62069 74.3163 640.658
1 8.96552 80.3805 720.653
1 9.31034 86.6825 807.044
1 9.65517 93.2224 900.078
1 10 100 1000
solution:
0.375158
-0.61321
0.257092
0.253129
approx_data_vector:
0.375158 0.204655 0.157564 0.296159 0.682711 1.37949 2.44878 3.95285 5.95396 8.51439 11.6964 15.5623 20.1744 25.5948 31.886 39.1101 47.3294 56.6062 67.0028 78.5815 91.4045 105.534 121.033 137.962 156.385 176.364 197.96 221.237 246.257 273.081
Press ENTER to continue...
- Eigen3
- matplot++
- simdjson
- boost ( boost::math if to be precise )
All of these dependencies are under or permissive licenses, or under academic licenses.
I don't care much about the code, so I guess you can say the code has a BSD-2 license, so, do almost whatever you want to do with this ( why you w)
- Get config file name from console
- Get data file name from console and make logic for it
- Make the code more generalized for other cases ( how ? in c++ there's no well done CAS, we don't have our SymPy like stuff, only some simple stuff. ). It needs research.
- If 3 is impossible, allow user to define their matrices to solve the system using config file or whatever else.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent