Algoritmos genéticos paralelizados
The objective of this research was to optimize the execution time of a genetic algorithm applied to the planning of agricultural programming using parallel computing techniques, tools such as Google Colab, Python 3.8, Torch library, where applied, it was not possible to parallelize all the modules i...
Furkejuvvon:
| Váldodahkki: | |
|---|---|
| Eará dahkkit: | |
| Materiálatiipa: | bachelorThesis |
| Giella: | spa |
| Almmustuhtton: |
2022
|
| Fáttát: | |
| Liŋkkat: | http://repositorio.espam.edu.ec/handle/42000/1934 |
| Fáddágilkorat: |
Lasit fáddágilkoriid
Eai fáddágilkorat, Lasit vuosttaš fáddágilkora!
|
Lasit fáddágilkoriid | Čoahkkáigeassu: | The objective of this research was to optimize the execution time of a genetic algorithm applied to the planning of agricultural programming using parallel computing techniques, tools such as Google Colab, Python 3.8, Torch library, where applied, it was not possible to parallelize all the modules in the coding phase, so the most complex in terms of number of operations had to be parallelized, among the results excellent improvements were obtained with times greater than 200% for solutions greater than 50 genes (Agricultural Products) compared to the none-parallelized algorithm. A web module was implemented to generate new instances, which received parameters such as: number of cycles for harvest, available area, date of start of cultivation, population (number of final solutions), number of products (number of products desired to be part of the solution), and this results in a population of solutions with the highest profit margins. |
|---|