I run Genetic Algorithm optimization for my study using Python often.
I cannot understand some settings for Genetic Algorithm optimization of MultiCharts, although I read this MultiCharts Manual (https://www.multicharts.com/trading-sof ... timization).
Does the "Set Population Size" means the Initialization Population size?
And what does the "number of individuals in population for crossover" mean?
For example, if "Set Population Size" is set to 200, and "number of individuals in population for crossover" is set to 2 (the MultiCharts default value), then the Initialization Population size would be 200, and in the next generation, only 2 of 200 would be crossover, while 198 of 200 would keep the same?
Any suggestions to setting "Set Population Size"? How many would be good for general cases?
Any proposed value for "number of individuals in population for crossover"? What number would be good for general cases?
What do some settings for Genetic Algorithm optimization mean?
- Polly MultiCharts
- Posts: 264
- Joined: Jul 20 2022
- Has thanked: 1 time
- Been thanked: 67 times
Hello shargo,
Set Population Size is the number of results in the optimization table. This population is initially calculated with random inputs.
The smaller the number of individuals in population for crossover, the more accurate the optimization result is, but the optimization will take longer.
The larger the number of individuals in population for crossover, the faster the optimization will be, but in the resulting table there is a greater probability of not getting the best results.
Set Population Size is the number of results in the optimization table. This population is initially calculated with random inputs.
The smaller the number of individuals in population for crossover, the more accurate the optimization result is, but the optimization will take longer.
The larger the number of individuals in population for crossover, the faster the optimization will be, but in the resulting table there is a greater probability of not getting the best results.