Curve Approximator

This tool helps with approximating a target curve from given datasets with a given precision.

Features

• Written in Go 1.20.2.
• Multithreading support.
• Around 1000x faster than an equivalent Python 3.10 script, measured on AMD Ryzen 7 5800X.

Getting started

1. Fill in your data in the Chart data sheet of Data.ods spreadsheet (you can find it in the Example folder).
2. Select all content of the Data for analysis sheet and paste it into Data.csv file. Make sure the delimiter is the tabulation character (\t). Optionally the decimal point can be represented by a comma. Make sure to save the file.
3. Install required dependencies.
4. Run the software without compiling using go run curve-approximator.go. After the calculation is done, the result will appear in the command line.
5. Fill in the calculated values into the Chart data sheet in the row starting with Optimal.
6. Optionally: Export the results to .pdf by printing the spreadsheet.

Compilation

To compile the software, run go build curve-approximator.go. After that, the resulting executable will be available under ./curve-approximator.

Flags

• -file - path to the file with data, ex. -file Example/Data.csv.
• -precision - precision to calculate the result to, ex. -precision 0.01 will calculate the optimal curve to a precision of 1%.
• -separator - decimal separator present in data, ex. -separator . will set . as the decimal separator.

Flags can be prefixed either with - and --.

Examples of usage:

# Run without compilation
go run curve-approximator.go -file Example/Data.csv -precision 0.001 -separator "."

# Run compiled executable
./curve-approximator --file Data.csv --precision 0.001 --separator ","