SimonBlanke /
Gradient-Free-Optimizers
| Total Complexity | 13 |
| Total Lines | 84 |
| Duplicated Lines | 44.05 % |
| Changes | 0 | ||
Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
| 1 | # Author: Simon Blanke |
||
| 2 | # Email: [email protected] |
||
| 3 | # License: MIT License |
||
| 4 | |||
| 5 | import numpy as np |
||
| 6 | |||
| 7 | from gradient_free_optimizers import ParticleSwarmOptimizer |
||
| 8 | |||
| 9 | n_iter = 100 |
||
| 10 | |||
| 11 | |||
| 12 | def get_score(pos_new): |
||
| 13 | return -(pos_new[0] * pos_new[0]) |
||
| 14 | |||
| 15 | |||
| 16 | space_dim = np.array([10]) |
||
| 17 | init_positions = [np.array([0]), np.array([1]), np.array([2]), np.array([3])] |
||
| 18 | |||
| 19 | |||
| 20 | View Code Duplication | def _base_test(opt, init_positions): |
|
|
Duplication
introduced
by
|
|||
| 21 | for nth_init in range(len(init_positions)): |
||
| 22 | pos_new = opt.init_pos(nth_init) |
||
| 23 | score_new = get_score(pos_new) |
||
| 24 | opt.evaluate(score_new) |
||
| 25 | |||
| 26 | for nth_iter in range(len(init_positions), n_iter): |
||
| 27 | pos_new = opt.iterate(nth_iter) |
||
| 28 | score_new = get_score(pos_new) |
||
| 29 | opt.evaluate(score_new) |
||
| 30 | |||
| 31 | |||
| 32 | def _test_ParticleSwarmOptimizer( |
||
| 33 | init_positions=init_positions, space_dim=space_dim, opt_para={} |
||
| 34 | ): |
||
| 35 | opt = ParticleSwarmOptimizer(init_positions, space_dim, opt_para) |
||
| 36 | _base_test(opt, init_positions) |
||
| 37 | |||
| 38 | |||
| 39 | View Code Duplication | def test_individuals(): |
|
| 40 | for init_positions in [ |
||
| 41 | [np.array([0])], |
||
| 42 | [np.array([0]), np.array([0])], |
||
| 43 | [np.array([0]), np.array([0])], |
||
| 44 | [ |
||
| 45 | np.array([0]), |
||
| 46 | np.array([0]), |
||
| 47 | np.array([0]), |
||
| 48 | np.array([0]), |
||
| 49 | np.array([0]), |
||
| 50 | np.array([0]), |
||
| 51 | np.array([0]), |
||
| 52 | np.array([0]), |
||
| 53 | np.array([0]), |
||
| 54 | np.array([0]), |
||
| 55 | np.array([0]), |
||
| 56 | np.array([0]), |
||
| 57 | np.array([0]), |
||
| 58 | np.array([0]), |
||
| 59 | np.array([0]), |
||
| 60 | np.array([0]), |
||
| 61 | np.array([0]), |
||
| 62 | np.array([0]), |
||
| 63 | ], |
||
| 64 | ]: |
||
| 65 | _test_ParticleSwarmOptimizer(init_positions) |
||
| 66 | |||
| 67 | |||
| 68 | def test_inertia(): |
||
| 69 | for inertia in [0.1, 0.9]: |
||
| 70 | opt_para = {"inertia": inertia} |
||
| 71 | _test_ParticleSwarmOptimizer(opt_para=opt_para) |
||
| 72 | |||
| 73 | |||
| 74 | def test_cognitive_weight(): |
||
| 75 | for cognitive_weight in [0.1, 0.9]: |
||
| 76 | opt_para = {"cognitive_weight": cognitive_weight} |
||
| 77 | _test_ParticleSwarmOptimizer(opt_para=opt_para) |
||
| 78 | |||
| 79 | |||
| 80 | def test_social_weight(): |
||
| 81 | for social_weight in [0.1, 0.9]: |
||
| 82 | opt_para = {"social_weight": social_weight} |
||
| 83 | _test_ParticleSwarmOptimizer(opt_para=opt_para) |
||
| 84 |
Loading history...