Rainbow-project
Project objective is to generate 14,000 random numbers that meet specific parameters. Numbers are generated using python script (numpy and csv libraries).
Parameters:
- Random numbers should fit into a table with 2,000 rows and 7 columns. Preferable in an Excel Sheet . No two rows can have the same set of seven numbers
- Random numbers must be from 1 to 50
- Numbers are arranged in ascending order from left to right, with the smallest number on the row on the extreme left column and largest number on the extreme right column
- The sum of the seven numbers in each row must be between 138 and 210 (both numbers included)
- The mean of the seven numbers in each row must be between 20 and 30 (Both numbers included)
- Difference between minimum number and maximum number in a row should be between 30 and 44 (Both numbers included)
- The 7th root of the product of all 7 numbers . Such that AxBxCxDxExFxG= KKK, therefore 7√KKK =KKK^(1/7). The 7th root of each row must be between 10.54 and 25.18 (these numbers are included)
- The standard deviation of the seven numbers in each row, must be between 11.0001 and 16.0001 (these numbers are included)
- We cannot repeat same number in the same row
- The value of the each number in a colums is represented in probabilityDistribution.csv (
- for column 'A' rows 1-14 in probabilityDistribution.csv,
- for column 'B' rows 15-36 in probabilityDistribution.csv,
- for column 'C' rows 37-65 in probabilityDistribution.csv,
- for column 'D' rows 66-92 in probabilityDistribution.csv,
- for column 'E' rows 93-118 in probabilityDistribution.csv,
- for column 'F' rows 119-140 in probabilityDistribution.csv,
- for column 'G' rows 141-155 in probabilityDistribution.csv
A detailed description of the project is in the file 'The Rainbow Project.pptx' Using pyinstaller I made executable files for windows (/build/data/data.exe) and linux (/dist/data / data). The Result.csv file is an example of an output file
Run program:
- Windows click on data.exe in /build/data
- Linux run terminal, go to /dist/data folder and then run command ./data
- Any OS - in terminal, go to folder with data.py file and then run command python data.py