Newer
Older
//
// Created by martin on 20.08.19.
//
void optimize_via_particle_swarm(l4n::NeuralNetwork& net,
l4n::ErrorFunction& ef) {
/* TRAINING METHOD SETUP */
std::vector<double> domain_bounds(2 * (net.get_n_weights() + net.get_n_biases()));
for (size_t i = 0; i < domain_bounds.size() / 2; ++i) {
domain_bounds[2 * i] = -10;
domain_bounds[2 * i + 1] = 10;
}
double c1 = 1.7;
double c2 = 1.7;
double w = 0.7;
size_t n_particles = 100;
size_t iter_max = 30;
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
/* if the maximal velocity from the previous step is less than 'gamma' times the current maximal velocity, then one
* terminating criterion is met */
double gamma = 0.5;
/* if 'delta' times 'n' particles are in the centroid neighborhood given by the radius 'epsilon', then the second
* terminating criterion is met ('n' is the total number of particles) */
double epsilon = 0.02;
double delta = 0.7;
l4n::ParticleSwarm swarm_01(
&domain_bounds,
c1,
c2,
w,
gamma,
epsilon,
delta,
n_particles,
iter_max
);
swarm_01.optimize(ef);
net.copy_parameter_space(swarm_01.get_parameters());
/* ERROR CALCULATION */
std::cout << "Run finished! Error of the network[Particle swarm]: " << ef.eval(nullptr) << std::endl;
std::cout
<< "***********************************************************************************************************************"
<< std::endl;
}
double optimize_via_gradient_descent(l4n::NeuralNetwork& net,
l4n::ErrorFunction& ef) {
std::cout
<< "***********************************************************************************************************************"
<< std::endl;
l4n::GradientDescentBB gd(1e-6,
1000);
gd.optimize(ef);
net.copy_parameter_space(gd.get_parameters());
/* ERROR CALCULATION */
double err = ef.eval(nullptr);
std::cout << "Run finished! Error of the network[Gradient descent]: " << err << std::endl;
/* Just for validation test purposes - NOT necessary for the example to work! */
return err;
}
double optimize_via_LBMQ(l4n::NeuralNetwork& net,
l4n::ErrorFunction& ef) {
size_t max_iterations = 10000;
size_t batch_size = 0;
double tolerance = 1e-6;
double tolerance_gradient = tolerance;
double tolerance_parameters = tolerance;
std::cout
<< "***********************************************************************************************************************"
<< std::endl;
l4n::LevenbergMarquardt lm(
max_iterations,
batch_size,
tolerance,
tolerance_gradient,
tolerance_parameters
);
lm.optimize(ef);
net.copy_parameter_space(lm.get_parameters());
/* ERROR CALCULATION */
double err = ef.eval(nullptr);
// std::cout << "Run finished! Error of the network[Levenberg-Marquardt]: " << err << std::endl;
/* Just for validation test purposes - NOT necessary for the example to work! */
return err;
}
int main() {
try{
/* Specify cutoff functions */
// l4n::CutoffFunction1 cutoff1(10.1);
l4n::CutoffFunction2 cutoff1(8);
// l4n::CutoffFunction2 cutoff2(15.2);
// l4n::CutoffFunction2 cutoff4(10.3);
// l4n::CutoffFunction2 cutoff5(12.9);
// l4n::CutoffFunction2 cutoff6(11);
/* Specify symmetry functions */
l4n::G1 sym_f1(&cutoff1);
l4n::G2 sym_f2(&cutoff1, 15, 8);
l4n::G2 sym_f3(&cutoff1, 10, 4);
// l4n::G3 sym_f4(&cutoff4, 0.3);
// l4n::G4 sym_f5(&cutoff5, 0.05, true, 0.05);
// l4n::G4 sym_f6(&cutoff5, 0.05, false, 0.05);
// l4n::G4 sym_f7(&cutoff6, 0.5, true, 0.05);
// l4n::G4 sym_f8(&cutoff6, 0.5, false, 0.05);
std::vector<l4n::SymmetryFunction*> helium_sym_funcs = {&sym_f1, &sym_f2, &sym_f3}; //, &sym_f4, &sym_f5, &sym_f6, &sym_f7, &sym_f8};
l4n::Element helium = l4n::Element("He",
helium_sym_funcs);
std::unordered_map<l4n::ELEMENT_SYMBOL, l4n::Element*> elements;
elements[l4n::ELEMENT_SYMBOL::He] = &helium;
/* Read data */
l4n::XYZReader reader("/home/martin/Desktop/HE21+T2.xyz");
reader.read();
std::cout << "Finished reading data" << std::endl;
std::shared_ptr<l4n::DataSet> ds = reader.get_acsf_data_set(elements);
/* Create a neural network */
std::unordered_map<l4n::ELEMENT_SYMBOL, std::vector<unsigned int>> n_hidden_neurons;
n_hidden_neurons[l4n::ELEMENT_SYMBOL::He] = {2, 1};
std::unordered_map<l4n::ELEMENT_SYMBOL, std::vector<l4n::NEURON_TYPE>> type_hidden_neurons;
type_hidden_neurons[l4n::ELEMENT_SYMBOL::He] = {l4n::NEURON_TYPE::LOGISTIC, l4n::NEURON_TYPE::LINEAR};
l4n::ACSFNeuralNetwork net(elements, *reader.get_element_list(), reader.contains_charge(), n_hidden_neurons, type_hidden_neurons);
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
// l4n::NeuralNetwork net;
// std::vector<std::shared_ptr<l4n::NeuronLinear>> inps;
// std::vector<size_t> inps_inds;
// for(unsigned int i = 0; i < 126; i++) {
// std::shared_ptr<l4n::NeuronLinear> inp = std::make_shared<l4n::NeuronLinear>();
// inps.emplace_back(inp);
// inps_inds.emplace_back(net.add_neuron(inp, l4n::BIAS_TYPE::NO_BIAS));
// }
//
// net.specify_input_neurons(inps_inds);
//
// std::vector<std::shared_ptr<l4n::NeuronLogistic>> hids;
//
// std::vector<unsigned int> hids_idxs;
// size_t idx;
// unsigned int n_hidden = 5;
// for(unsigned int i = 0; i < n_hidden; i++) {
// std::shared_ptr<l4n::NeuronLogistic> hid = std::make_shared<l4n::NeuronLogistic>();
// hids.emplace_back(hid);
// idx = net.add_neuron(hid, l4n::BIAS_TYPE::NEXT_BIAS);
// hids_idxs.emplace_back(idx);
//
// for(unsigned int j = 0; j < 126; j++) {
// net.add_connection_simple(j, idx);
// }
// }
//
// std::shared_ptr<l4n::NeuronLinear> out = std::make_shared<l4n::NeuronLinear>();
// idx = net.add_neuron(out, l4n::BIAS_TYPE::NO_BIAS);
// std::vector<size_t> out_inds = {idx};
// for(unsigned int i = 0; i < n_hidden; i++) {
// net.add_connection_simple(hids_idxs.at(i), idx);
// }
// net.specify_output_neurons(out_inds);
l4n::MSE mse(&net, ds.get());
net.randomize_parameters();
// optimize_via_particle_swarm(net, mse);
double err1 = optimize_via_LBMQ(net, mse);
double err2 = optimize_via_gradient_descent(net, mse);
if(err2 > 0.00001) {
throw std::runtime_error("Training was incorrect!");
}
/* Print fit comparison with real data */
std::vector<double> output;
output.resize(1);
for(auto e : *ds->get_data()) {
for(unsigned int i = 0; i < e.first.size(); i++) {
std::cout << e.first.at(i) << " ";
if(i % 3 == 2) {
std::cout << std::endl;
}
}
std::cout << e.second.at(0) << " ";
net.eval_single(e.first, output);
std::cout << output.at(0) << std::endl;
}
} catch (const std::exception& e) {
std::cerr << e.what() << std::endl;
exit(EXIT_FAILURE);
}
return 0;