Relintai/pmlpp
1
0
Fork 0
mirror of https://github.com/Relintai/pmlpp.git synced 2025-04-17 03:46:04 +02:00
Code Issues Packages Projects Releases Wiki Activity

Codestyle for MLPPANN.

Browse Source
This commit is contained in:
Relintai 2023-02-13 00:56:09 +01:00
parent 3feed886c4
commit e25892aacf
2 changed files with 319 additions and 316 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

605
mlpp/ann/ann.cpp
View File

@ -16,35 +16,35 @@
#include <random> #include <random>
std::vector<real_t> MLPPANN::model_set_test(std::vector<std::vector<real_t>> X) { std::vector<real_t> MLPPANN::model_set_test(std::vector<std::vector<real_t>> X) {
if (!network.empty()) { if (!_network.empty()) {
network[0].input = X; _network[0].input = X;
network[0].forwardPass(); _network[0].forwardPass();
for (uint32_t i = 1; i < network.size(); i++) { for (uint32_t i = 1; i < _network.size(); i++) {
network[i].input = network[i - 1].a; _network[i].input = _network[i - 1].a;
network[i].forwardPass(); _network[i].forwardPass();
} }
outputLayer->input = network[network.size() - 1].a; _output_layer->input = _network[_network.size() - 1].a;
} else { } else {
outputLayer->input = X; _output_layer->input = X;
} }
outputLayer->forwardPass(); _output_layer->forwardPass();
return outputLayer->a; return _output_layer->a;
} }
real_t MLPPANN::model_test(std::vector<real_t> x) { real_t MLPPANN::model_test(std::vector<real_t> x) {
if (!network.empty()) { if (!_network.empty()) {
network[0].Test(x); _network[0].Test(x);
for (uint32_t i = 1; i < network.size(); i++) { for (uint32_t i = 1; i < _network.size(); i++) {
network[i].Test(network[i - 1].a_test); _network[i].Test(_network[i - 1].a_test);
} }
outputLayer->Test(network[network.size() - 1].a_test); _output_layer->Test(_network[_network.size() - 1].a_test);
} else { } else {
outputLayer->Test(x); _output_layer->Test(x);
} }
return outputLayer->a_test; return _output_layer->a_test;
} }
void MLPPANN::gradient_descent(real_t learning_rate, int max_epoch, bool ui) { void MLPPANN::gradient_descent(real_t learning_rate, int max_epoch, bool ui) {
@ -57,26 +57,26 @@ void MLPPANN::gradient_descent(real_t learning_rate, int max_epoch, bool ui) {
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
alg.printMatrix(network[network.size() - 1].weights); alg.printMatrix(_network[_network.size() - 1].weights);
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
cost_prev = cost(y_hat, outputSet); cost_prev = cost(_y_hat, _output_set);
auto grads = compute_gradients(y_hat, outputSet); auto grads = compute_gradients(_y_hat, _output_set);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
cumulativeHiddenLayerWGrad = alg.scalarMultiply(learning_rate / n, cumulativeHiddenLayerWGrad); cumulative_hidden_layer_w_grad = alg.scalarMultiply(learning_rate / _n, cumulative_hidden_layer_w_grad);
outputWGrad = alg.scalarMultiply(learning_rate / n, outputWGrad); output_w_grad = alg.scalarMultiply(learning_rate / _n, output_w_grad);
update_parameters(cumulativeHiddenLayerWGrad, outputWGrad, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(cumulative_hidden_layer_w_grad, output_w_grad, learning_rate); // subject to change. may want bias to have this matrix too.
std::cout << learning_rate << std::endl; std::cout << learning_rate << std::endl;
forward_pass(); forward_pass();
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputSet); print_ui(epoch, cost_prev, _y_hat, _output_set);
} }
epoch++; epoch++;
@ -95,28 +95,28 @@ void MLPPANN::sgd(real_t learning_rate, int max_epoch, bool ui) {
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
std::random_device rd; std::random_device rd;
std::default_random_engine generator(rd()); std::default_random_engine generator(rd());
std::uniform_int_distribution<int> distribution(0, int(n - 1)); std::uniform_int_distribution<int> distribution(0, int(_n - 1));
int outputIndex = distribution(generator); int outputIndex = distribution(generator);
std::vector<real_t> y_hat = model_set_test({ inputSet[outputIndex] }); std::vector<real_t> y_hat = model_set_test({ _input_set[outputIndex] });
cost_prev = cost({ y_hat }, { outputSet[outputIndex] }); cost_prev = cost({ y_hat }, { _output_set[outputIndex] });
auto grads = compute_gradients(y_hat, { outputSet[outputIndex] }); auto grads = compute_gradients(y_hat, { _output_set[outputIndex] });
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
cumulativeHiddenLayerWGrad = alg.scalarMultiply(learning_rate / n, cumulativeHiddenLayerWGrad); cumulative_hidden_layer_w_grad = alg.scalarMultiply(learning_rate / _n, cumulative_hidden_layer_w_grad);
outputWGrad = alg.scalarMultiply(learning_rate / n, outputWGrad); output_w_grad = alg.scalarMultiply(learning_rate / _n, output_w_grad);
update_parameters(cumulativeHiddenLayerWGrad, outputWGrad, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(cumulative_hidden_layer_w_grad, output_w_grad, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test({ inputSet[outputIndex] }); y_hat = model_set_test({ _input_set[outputIndex] });
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, { outputSet[outputIndex] }); print_ui(epoch, cost_prev, y_hat, { _output_set[outputIndex] });
} }
epoch++; epoch++;
@ -137,33 +137,33 @@ void MLPPANN::mbgd(real_t learning_rate, int max_epoch, int mini_batch_size, boo
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
cumulativeHiddenLayerWGrad = alg.scalarMultiply(learning_rate / n, cumulativeHiddenLayerWGrad); cumulative_hidden_layer_w_grad = alg.scalarMultiply(learning_rate / _n, cumulative_hidden_layer_w_grad);
outputWGrad = alg.scalarMultiply(learning_rate / n, outputWGrad); output_w_grad = alg.scalarMultiply(learning_rate / _n, output_w_grad);
update_parameters(cumulativeHiddenLayerWGrad, outputWGrad, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(cumulative_hidden_layer_w_grad, output_w_grad, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
@ -186,50 +186,50 @@ void MLPPANN::momentum(real_t learning_rate, int max_epoch, int mini_batch_size,
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> v_hidden; std::vector<std::vector<std::vector<real_t>>> v_hidden;
std::vector<real_t> v_output; std::vector<real_t> v_output;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && v_hidden.empty()) { // Initing our tensor if (!_network.empty() && v_hidden.empty()) { // Initing our tensor
v_hidden = alg.resize(v_hidden, cumulativeHiddenLayerWGrad); v_hidden = alg.resize(v_hidden, cumulative_hidden_layer_w_grad);
} }
if (v_output.empty()) { if (v_output.empty()) {
v_output.resize(outputWGrad.size()); v_output.resize(output_w_grad.size());
} }
if (nag) { // "Aposterori" calculation if (nag) { // "Aposterori" calculation
update_parameters(v_hidden, v_output, 0); // DON'T update bias. update_parameters(v_hidden, v_output, 0); // DON'T update bias.
} }
v_hidden = alg.addition(alg.scalarMultiply(gamma, v_hidden), alg.scalarMultiply(learning_rate / n, cumulativeHiddenLayerWGrad)); v_hidden = alg.addition(alg.scalarMultiply(gamma, v_hidden), alg.scalarMultiply(learning_rate / _n, cumulative_hidden_layer_w_grad));
v_output = alg.addition(alg.scalarMultiply(gamma, v_output), alg.scalarMultiply(learning_rate / n, outputWGrad)); v_output = alg.addition(alg.scalarMultiply(gamma, v_output), alg.scalarMultiply(learning_rate / _n, output_w_grad));
update_parameters(v_hidden, v_output, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(v_hidden, v_output, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
@ -252,49 +252,49 @@ void MLPPANN::adagrad(real_t learning_rate, int max_epoch, int mini_batch_size,
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> v_hidden; std::vector<std::vector<std::vector<real_t>>> v_hidden;
std::vector<real_t> v_output; std::vector<real_t> v_output;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && v_hidden.empty()) { // Initing our tensor if (!_network.empty() && v_hidden.empty()) { // Initing our tensor
v_hidden = alg.resize(v_hidden, cumulativeHiddenLayerWGrad); v_hidden = alg.resize(v_hidden, cumulative_hidden_layer_w_grad);
} }
if (v_output.empty()) { if (v_output.empty()) {
v_output.resize(outputWGrad.size()); v_output.resize(output_w_grad.size());
} }
v_hidden = alg.addition(v_hidden, alg.exponentiate(cumulativeHiddenLayerWGrad, 2)); v_hidden = alg.addition(v_hidden, alg.exponentiate(cumulative_hidden_layer_w_grad, 2));
v_output = alg.addition(v_output, alg.exponentiate(outputWGrad, 2)); v_output = alg.addition(v_output, alg.exponentiate(output_w_grad, 2));
std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(cumulativeHiddenLayerWGrad, alg.scalarAdd(e, alg.sqrt(v_hidden)))); std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(cumulative_hidden_layer_w_grad, alg.scalarAdd(e, alg.sqrt(v_hidden))));
std::vector<real_t> outputLayerUpdation = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(outputWGrad, alg.scalarAdd(e, alg.sqrt(v_output)))); std::vector<real_t> output_layer_updation = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(output_w_grad, alg.scalarAdd(e, alg.sqrt(v_output))));
update_parameters(hiddenLayerUpdations, outputLayerUpdation, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(hidden_layer_updations, output_layer_updation, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
epoch++; epoch++;
@ -315,55 +315,58 @@ void MLPPANN::adadelta(real_t learning_rate, int max_epoch, int mini_batch_size,
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> v_hidden; std::vector<std::vector<std::vector<real_t>>> v_hidden;
std::vector<real_t> v_output; std::vector<real_t> v_output;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && v_hidden.empty()) { // Initing our tensor if (!_network.empty() && v_hidden.empty()) { // Initing our tensor
v_hidden = alg.resize(v_hidden, cumulativeHiddenLayerWGrad); v_hidden = alg.resize(v_hidden, cumulative_hidden_layer_w_grad);
} }
if (v_output.empty()) { if (v_output.empty()) {
v_output.resize(outputWGrad.size()); v_output.resize(output_w_grad.size());
} }
v_hidden = alg.addition(alg.scalarMultiply(1 - b1, v_hidden), alg.scalarMultiply(b1, alg.exponentiate(cumulativeHiddenLayerWGrad, 2))); v_hidden = alg.addition(alg.scalarMultiply(1 - b1, v_hidden), alg.scalarMultiply(b1, alg.exponentiate(cumulative_hidden_layer_w_grad, 2)));
v_output = alg.addition(v_output, alg.exponentiate(outputWGrad, 2)); v_output = alg.addition(v_output, alg.exponentiate(output_w_grad, 2));
std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(cumulativeHiddenLayerWGrad, alg.scalarAdd(e, alg.sqrt(v_hidden)))); std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(cumulative_hidden_layer_w_grad, alg.scalarAdd(e, alg.sqrt(v_hidden))));
std::vector<real_t> outputLayerUpdation = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(outputWGrad, alg.scalarAdd(e, alg.sqrt(v_output)))); std::vector<real_t> output_layer_updation = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(output_w_grad, alg.scalarAdd(e, alg.sqrt(v_output))));
update_parameters(hiddenLayerUpdations, outputLayerUpdation, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(hidden_layer_updations, output_layer_updation, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
epoch++; epoch++;
if (epoch > max_epoch) { if (epoch > max_epoch) {
break; break;
} }
} }
forward_pass(); forward_pass();
} }
@ -376,13 +379,13 @@ void MLPPANN::adam(real_t learning_rate, int max_epoch, int mini_batch_size, rea
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> m_hidden; std::vector<std::vector<std::vector<real_t>>> m_hidden;
@ -391,30 +394,30 @@ void MLPPANN::adam(real_t learning_rate, int max_epoch, int mini_batch_size, rea
std::vector<real_t> m_output; std::vector<real_t> m_output;
std::vector<real_t> v_output; std::vector<real_t> v_output;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && m_hidden.empty() && v_hidden.empty()) { // Initing our tensor if (!_network.empty() && m_hidden.empty() && v_hidden.empty()) { // Initing our tensor
m_hidden = alg.resize(m_hidden, cumulativeHiddenLayerWGrad); m_hidden = alg.resize(m_hidden, cumulative_hidden_layer_w_grad);
v_hidden = alg.resize(v_hidden, cumulativeHiddenLayerWGrad); v_hidden = alg.resize(v_hidden, cumulative_hidden_layer_w_grad);
} }
if (m_output.empty() && v_output.empty()) { if (m_output.empty() && v_output.empty()) {
m_output.resize(outputWGrad.size()); m_output.resize(output_w_grad.size());
v_output.resize(outputWGrad.size()); v_output.resize(output_w_grad.size());
} }
m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulativeHiddenLayerWGrad)); m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulative_hidden_layer_w_grad));
v_hidden = alg.addition(alg.scalarMultiply(b2, v_hidden), alg.scalarMultiply(1 - b2, alg.exponentiate(cumulativeHiddenLayerWGrad, 2))); v_hidden = alg.addition(alg.scalarMultiply(b2, v_hidden), alg.scalarMultiply(1 - b2, alg.exponentiate(cumulative_hidden_layer_w_grad, 2)));
m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, outputWGrad)); m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, output_w_grad));
v_output = alg.addition(alg.scalarMultiply(b2, v_output), alg.scalarMultiply(1 - b2, alg.exponentiate(outputWGrad, 2))); v_output = alg.addition(alg.scalarMultiply(b2, v_output), alg.scalarMultiply(1 - b2, alg.exponentiate(output_w_grad, 2)));
std::vector<std::vector<std::vector<real_t>>> m_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_hidden); std::vector<std::vector<std::vector<real_t>>> m_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_hidden);
std::vector<std::vector<std::vector<real_t>>> v_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_hidden); std::vector<std::vector<std::vector<real_t>>> v_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_hidden);
@ -422,14 +425,14 @@ void MLPPANN::adam(real_t learning_rate, int max_epoch, int mini_batch_size, rea
std::vector<real_t> m_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_output); std::vector<real_t> m_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_output);
std::vector<real_t> v_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_output); std::vector<real_t> v_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_output);
std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_hidden_hat, alg.scalarAdd(e, alg.sqrt(v_hidden_hat)))); std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_hidden_hat, alg.scalarAdd(e, alg.sqrt(v_hidden_hat))));
std::vector<real_t> outputLayerUpdation = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_output_hat, alg.scalarAdd(e, alg.sqrt(v_output_hat)))); std::vector<real_t> output_layer_updation = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_output_hat, alg.scalarAdd(e, alg.sqrt(v_output_hat))));
update_parameters(hiddenLayerUpdations, outputLayerUpdation, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(hidden_layer_updations, output_layer_updation, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
@ -451,13 +454,13 @@ void MLPPANN::adamax(real_t learning_rate, int max_epoch, int mini_batch_size, r
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> m_hidden; std::vector<std::vector<std::vector<real_t>>> m_hidden;
@ -466,43 +469,43 @@ void MLPPANN::adamax(real_t learning_rate, int max_epoch, int mini_batch_size, r
std::vector<real_t> m_output; std::vector<real_t> m_output;
std::vector<real_t> u_output; std::vector<real_t> u_output;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && m_hidden.empty() && u_hidden.empty()) { // Initing our tensor if (!_network.empty() && m_hidden.empty() && u_hidden.empty()) { // Initing our tensor
m_hidden = alg.resize(m_hidden, cumulativeHiddenLayerWGrad); m_hidden = alg.resize(m_hidden, cumulative_hidden_layer_w_grad);
u_hidden = alg.resize(u_hidden, cumulativeHiddenLayerWGrad); u_hidden = alg.resize(u_hidden, cumulative_hidden_layer_w_grad);
} }
if (m_output.empty() && u_output.empty()) { if (m_output.empty() && u_output.empty()) {
m_output.resize(outputWGrad.size()); m_output.resize(output_w_grad.size());
u_output.resize(outputWGrad.size()); u_output.resize(output_w_grad.size());
} }
m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulativeHiddenLayerWGrad)); m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulative_hidden_layer_w_grad));
u_hidden = alg.max(alg.scalarMultiply(b2, u_hidden), alg.abs(cumulativeHiddenLayerWGrad)); u_hidden = alg.max(alg.scalarMultiply(b2, u_hidden), alg.abs(cumulative_hidden_layer_w_grad));
m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, outputWGrad)); m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, output_w_grad));
u_output = alg.max(alg.scalarMultiply(b2, u_output), alg.abs(outputWGrad)); u_output = alg.max(alg.scalarMultiply(b2, u_output), alg.abs(output_w_grad));
std::vector<std::vector<std::vector<real_t>>> m_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_hidden); std::vector<std::vector<std::vector<real_t>>> m_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_hidden);
std::vector<real_t> m_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_output); std::vector<real_t> m_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_output);
std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_hidden_hat, alg.scalarAdd(e, u_hidden))); std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_hidden_hat, alg.scalarAdd(e, u_hidden)));
std::vector<real_t> outputLayerUpdation = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_output_hat, alg.scalarAdd(e, u_output))); std::vector<real_t> output_layer_updation = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_output_hat, alg.scalarAdd(e, u_output)));
update_parameters(hiddenLayerUpdations, outputLayerUpdation, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(hidden_layer_updations, output_layer_updation, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
@ -524,13 +527,13 @@ void MLPPANN::nadam(real_t learning_rate, int max_epoch, int mini_batch_size, re
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> m_hidden; std::vector<std::vector<std::vector<real_t>>> m_hidden;
@ -539,47 +542,47 @@ void MLPPANN::nadam(real_t learning_rate, int max_epoch, int mini_batch_size, re
std::vector<real_t> m_output; std::vector<real_t> m_output;
std::vector<real_t> v_output; std::vector<real_t> v_output;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && m_hidden.empty() && v_hidden.empty()) { // Initing our tensor if (!_network.empty() && m_hidden.empty() && v_hidden.empty()) { // Initing our tensor
m_hidden = alg.resize(m_hidden, cumulativeHiddenLayerWGrad); m_hidden = alg.resize(m_hidden, cumulative_hidden_layer_w_grad);
v_hidden = alg.resize(v_hidden, cumulativeHiddenLayerWGrad); v_hidden = alg.resize(v_hidden, cumulative_hidden_layer_w_grad);
} }
if (m_output.empty() && v_output.empty()) { if (m_output.empty() && v_output.empty()) {
m_output.resize(outputWGrad.size()); m_output.resize(output_w_grad.size());
v_output.resize(outputWGrad.size()); v_output.resize(output_w_grad.size());
} }
m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulativeHiddenLayerWGrad)); m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulative_hidden_layer_w_grad));
v_hidden = alg.addition(alg.scalarMultiply(b2, v_hidden), alg.scalarMultiply(1 - b2, alg.exponentiate(cumulativeHiddenLayerWGrad, 2))); v_hidden = alg.addition(alg.scalarMultiply(b2, v_hidden), alg.scalarMultiply(1 - b2, alg.exponentiate(cumulative_hidden_layer_w_grad, 2)));
m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, outputWGrad)); m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, output_w_grad));
v_output = alg.addition(alg.scalarMultiply(b2, v_output), alg.scalarMultiply(1 - b2, alg.exponentiate(outputWGrad, 2))); v_output = alg.addition(alg.scalarMultiply(b2, v_output), alg.scalarMultiply(1 - b2, alg.exponentiate(output_w_grad, 2)));
std::vector<std::vector<std::vector<real_t>>> m_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_hidden); std::vector<std::vector<std::vector<real_t>>> m_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_hidden);
std::vector<std::vector<std::vector<real_t>>> v_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_hidden); std::vector<std::vector<std::vector<real_t>>> v_hidden_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_hidden);
std::vector<std::vector<std::vector<real_t>>> m_hidden_final = alg.addition(alg.scalarMultiply(b1, m_hidden_hat), alg.scalarMultiply((1 - b1) / (1 - std::pow(b1, epoch)), cumulativeHiddenLayerWGrad)); std::vector<std::vector<std::vector<real_t>>> m_hidden_final = alg.addition(alg.scalarMultiply(b1, m_hidden_hat), alg.scalarMultiply((1 - b1) / (1 - std::pow(b1, epoch)), cumulative_hidden_layer_w_grad));
std::vector<real_t> m_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_output); std::vector<real_t> m_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b1, epoch)), m_output);
std::vector<real_t> v_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_output); std::vector<real_t> v_output_hat = alg.scalarMultiply(1 / (1 - std::pow(b2, epoch)), v_output);
std::vector<real_t> m_output_final = alg.addition(alg.scalarMultiply(b1, m_output_hat), alg.scalarMultiply((1 - b1) / (1 - std::pow(b1, epoch)), outputWGrad)); std::vector<real_t> m_output_final = alg.addition(alg.scalarMultiply(b1, m_output_hat), alg.scalarMultiply((1 - b1) / (1 - std::pow(b1, epoch)), output_w_grad));
std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_hidden_final, alg.scalarAdd(e, alg.sqrt(v_hidden_hat)))); std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_hidden_final, alg.scalarAdd(e, alg.sqrt(v_hidden_hat))));
std::vector<real_t> outputLayerUpdation = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_output_final, alg.scalarAdd(e, alg.sqrt(v_output_hat)))); std::vector<real_t> output_layer_updation = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_output_final, alg.scalarAdd(e, alg.sqrt(v_output_hat))));
update_parameters(hiddenLayerUpdations, outputLayerUpdation, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(hidden_layer_updations, output_layer_updation, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
@ -602,13 +605,13 @@ void MLPPANN::amsgrad(real_t learning_rate, int max_epoch, int mini_batch_size,
real_t initial_learning_rate = learning_rate; real_t initial_learning_rate = learning_rate;
// Creating the mini-batches // Creating the mini-batches
int n_mini_batch = n / mini_batch_size; int n_mini_batch = _n / mini_batch_size;
// always evaluate the result // always evaluate the result
// always do forward pass only ONCE at end. // always do forward pass only ONCE at end.
auto batches = MLPPUtilities::createMiniBatches(inputSet, outputSet, n_mini_batch); auto batches = MLPPUtilities::createMiniBatches(_input_set, _output_set, n_mini_batch);
auto inputMiniBatches = std::get<0>(batches); auto input_mini_batches = std::get<0>(batches);
auto outputMiniBatches = std::get<1>(batches); auto output_mini_batches = std::get<1>(batches);
// Initializing necessary components for Adam. // Initializing necessary components for Adam.
std::vector<std::vector<std::vector<real_t>>> m_hidden; std::vector<std::vector<std::vector<real_t>>> m_hidden;
@ -621,45 +624,45 @@ void MLPPANN::amsgrad(real_t learning_rate, int max_epoch, int mini_batch_size,
std::vector<real_t> v_output_hat; std::vector<real_t> v_output_hat;
while (true) { while (true) {
learning_rate = apply_learning_rate_scheduler(initial_learning_rate, decayConstant, epoch, dropRate); learning_rate = apply_learning_rate_scheduler(initial_learning_rate, _decay_constant, epoch, _drop_rate);
for (int i = 0; i < n_mini_batch; i++) { for (int i = 0; i < n_mini_batch; i++) {
std::vector<real_t> y_hat = model_set_test(inputMiniBatches[i]); std::vector<real_t> y_hat = model_set_test(input_mini_batches[i]);
cost_prev = cost(y_hat, outputMiniBatches[i]); cost_prev = cost(y_hat, output_mini_batches[i]);
auto grads = compute_gradients(y_hat, outputMiniBatches[i]); auto grads = compute_gradients(y_hat, output_mini_batches[i]);
auto cumulativeHiddenLayerWGrad = std::get<0>(grads); auto cumulative_hidden_layer_w_grad = std::get<0>(grads);
auto outputWGrad = std::get<1>(grads); auto output_w_grad = std::get<1>(grads);
if (!network.empty() && m_hidden.empty() && v_hidden.empty()) { // Initing our tensor if (!_network.empty() && m_hidden.empty() && v_hidden.empty()) { // Initing our tensor
m_hidden = alg.resize(m_hidden, cumulativeHiddenLayerWGrad); m_hidden = alg.resize(m_hidden, cumulative_hidden_layer_w_grad);
v_hidden = alg.resize(v_hidden, cumulativeHiddenLayerWGrad); v_hidden = alg.resize(v_hidden, cumulative_hidden_layer_w_grad);
v_hidden_hat = alg.resize(v_hidden_hat, cumulativeHiddenLayerWGrad); v_hidden_hat = alg.resize(v_hidden_hat, cumulative_hidden_layer_w_grad);
} }
if (m_output.empty() && v_output.empty()) { if (m_output.empty() && v_output.empty()) {
m_output.resize(outputWGrad.size()); m_output.resize(output_w_grad.size());
v_output.resize(outputWGrad.size()); v_output.resize(output_w_grad.size());
v_output_hat.resize(outputWGrad.size()); v_output_hat.resize(output_w_grad.size());
} }
m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulativeHiddenLayerWGrad)); m_hidden = alg.addition(alg.scalarMultiply(b1, m_hidden), alg.scalarMultiply(1 - b1, cumulative_hidden_layer_w_grad));
v_hidden = alg.addition(alg.scalarMultiply(b2, v_hidden), alg.scalarMultiply(1 - b2, alg.exponentiate(cumulativeHiddenLayerWGrad, 2))); v_hidden = alg.addition(alg.scalarMultiply(b2, v_hidden), alg.scalarMultiply(1 - b2, alg.exponentiate(cumulative_hidden_layer_w_grad, 2)));
m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, outputWGrad)); m_output = alg.addition(alg.scalarMultiply(b1, m_output), alg.scalarMultiply(1 - b1, output_w_grad));
v_output = alg.addition(alg.scalarMultiply(b2, v_output), alg.scalarMultiply(1 - b2, alg.exponentiate(outputWGrad, 2))); v_output = alg.addition(alg.scalarMultiply(b2, v_output), alg.scalarMultiply(1 - b2, alg.exponentiate(output_w_grad, 2)));
v_hidden_hat = alg.max(v_hidden_hat, v_hidden); v_hidden_hat = alg.max(v_hidden_hat, v_hidden);
v_output_hat = alg.max(v_output_hat, v_output); v_output_hat = alg.max(v_output_hat, v_output);
std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_hidden, alg.scalarAdd(e, alg.sqrt(v_hidden_hat)))); std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_hidden, alg.scalarAdd(e, alg.sqrt(v_hidden_hat))));
std::vector<real_t> outputLayerUpdation = alg.scalarMultiply(learning_rate / n, alg.elementWiseDivision(m_output, alg.scalarAdd(e, alg.sqrt(v_output_hat)))); std::vector<real_t> output_layer_updation = alg.scalarMultiply(learning_rate / _n, alg.elementWiseDivision(m_output, alg.scalarAdd(e, alg.sqrt(v_output_hat))));
update_parameters(hiddenLayerUpdations, outputLayerUpdation, learning_rate); // subject to change. may want bias to have this matrix too. update_parameters(hidden_layer_updations, output_layer_updation, learning_rate); // subject to change. may want bias to have this matrix too.
y_hat = model_set_test(inputMiniBatches[i]); y_hat = model_set_test(input_mini_batches[i]);
if (ui) { if (ui) {
print_ui(epoch, cost_prev, y_hat, outputMiniBatches[i]); print_ui(epoch, cost_prev, y_hat, output_mini_batches[i]);
} }
} }
@ -676,82 +679,82 @@ void MLPPANN::amsgrad(real_t learning_rate, int max_epoch, int mini_batch_size,
real_t MLPPANN::score() { real_t MLPPANN::score() {
MLPPUtilities util; MLPPUtilities util;
forward_pass(); forward_pass();
return util.performance(y_hat, outputSet); return util.performance(_y_hat, _output_set);
} }
void MLPPANN::save(std::string fileName) { void MLPPANN::save(std::string fileName) {
MLPPUtilities util; MLPPUtilities util;
if (!network.empty()) { if (!_network.empty()) {
util.saveParameters(fileName, network[0].weights, network[0].bias, false, 1); util.saveParameters(fileName, _network[0].weights, _network[0].bias, false, 1);
for (uint32_t i = 1; i < network.size(); i++) { for (uint32_t i = 1; i < _network.size(); i++) {
util.saveParameters(fileName, network[i].weights, network[i].bias, true, i + 1); util.saveParameters(fileName, _network[i].weights, _network[i].bias, true, i + 1);
} }
util.saveParameters(fileName, outputLayer->weights, outputLayer->bias, true, network.size() + 1); util.saveParameters(fileName, _output_layer->weights, _output_layer->bias, true, _network.size() + 1);
} else { } else {
util.saveParameters(fileName, outputLayer->weights, outputLayer->bias, false, network.size() + 1); util.saveParameters(fileName, _output_layer->weights, _output_layer->bias, false, _network.size() + 1);
} }
} }
void MLPPANN::set_learning_rate_scheduler(std::string type, real_t decayConstant) { void MLPPANN::set_learning_rate_scheduler(std::string type, real_t decay_constant) {
lrScheduler = type; _lr_scheduler = type;
MLPPANN::decayConstant = decayConstant; _decay_constant = decay_constant;
} }
void MLPPANN::set_learning_rate_scheduler_drop(std::string type, real_t decayConstant, real_t dropRate) { void MLPPANN::set_learning_rate_scheduler_drop(std::string type, real_t decay_constant, real_t drop_rate) {
lrScheduler = type; _lr_scheduler = type;
MLPPANN::decayConstant = decayConstant; _decay_constant = decay_constant;
MLPPANN::dropRate = dropRate; _drop_rate = drop_rate;
} }
// https://en.wikipedia.org/wiki/Learning_rate // https://en.wikipedia.org/wiki/Learning_rate
// Learning Rate Decay (C2W2L09) - Andrew Ng - Deep Learning Specialization // Learning Rate Decay (C2W2L09) - Andrew Ng - Deep Learning Specialization
real_t MLPPANN::apply_learning_rate_scheduler(real_t learningRate, real_t decayConstant, real_t epoch, real_t dropRate) { real_t MLPPANN::apply_learning_rate_scheduler(real_t learning_rate, real_t decay_constant, real_t epoch, real_t drop_rate) {
if (lrScheduler == "Time") { if (_lr_scheduler == "Time") {
return learningRate / (1 + decayConstant * epoch); return learning_rate / (1 + decay_constant * epoch);
} else if (lrScheduler == "Epoch") { } else if (_lr_scheduler == "Epoch") {
return learningRate * (decayConstant / std::sqrt(epoch)); return learning_rate * (decay_constant / std::sqrt(epoch));
} else if (lrScheduler == "Step") { } else if (_lr_scheduler == "Step") {
return learningRate * std::pow(decayConstant, int((1 + epoch) / dropRate)); // Utilizing an explicit int conversion implicitly takes the floor. return learning_rate * std::pow(decay_constant, int((1 + epoch) / drop_rate)); // Utilizing an explicit int conversion implicitly takes the floor.
} else if (lrScheduler == "Exponential") { } else if (_lr_scheduler == "Exponential") {
return learningRate * std::exp(-decayConstant * epoch); return learning_rate * std::exp(-decay_constant * epoch);
} }
return learningRate; return learning_rate;
} }
void MLPPANN::add_layer(int n_hidden, std::string activation, std::string weightInit, std::string reg, real_t lambda, real_t alpha) { void MLPPANN::add_layer(int n_hidden, std::string activation, std::string weightInit, std::string reg, real_t lambda, real_t alpha) {
if (network.empty()) { if (_network.empty()) {
network.push_back(MLPPOldHiddenLayer(n_hidden, activation, inputSet, weightInit, reg, lambda, alpha)); _network.push_back(MLPPOldHiddenLayer(n_hidden, activation, _input_set, weightInit, reg, lambda, alpha));
network[0].forwardPass(); _network[0].forwardPass();
} else { } else {
network.push_back(MLPPOldHiddenLayer(n_hidden, activation, network[network.size() - 1].a, weightInit, reg, lambda, alpha)); _network.push_back(MLPPOldHiddenLayer(n_hidden, activation, _network[_network.size() - 1].a, weightInit, reg, lambda, alpha));
network[network.size() - 1].forwardPass(); _network[_network.size() - 1].forwardPass();
} }
} }
void MLPPANN::add_output_layer(std::string activation, std::string loss, std::string weightInit, std::string reg, real_t lambda, real_t alpha) { void MLPPANN::add_output_layer(std::string activation, std::string loss, std::string weightInit, std::string reg, real_t lambda, real_t alpha) {
if (!network.empty()) { if (!_network.empty()) {
outputLayer = new MLPPOldOutputLayer(network[network.size() - 1].n_hidden, activation, loss, network[network.size() - 1].a, weightInit, reg, lambda, alpha); _output_layer = new MLPPOldOutputLayer(_network[_network.size() - 1].n_hidden, activation, loss, _network[_network.size() - 1].a, weightInit, reg, lambda, alpha);
} else { } else {
outputLayer = new MLPPOldOutputLayer(k, activation, loss, inputSet, weightInit, reg, lambda, alpha); _output_layer = new MLPPOldOutputLayer(_k, activation, loss, _input_set, weightInit, reg, lambda, alpha);
} }
} }
MLPPANN::MLPPANN(std::vector<std::vector<real_t>> p_inputSet, std::vector<real_t> p_outputSet) { MLPPANN::MLPPANN(std::vector<std::vector<real_t>> p_input_set, std::vector<real_t> p_output_set) {
inputSet = p_inputSet; _input_set = p_input_set;
outputSet = p_outputSet; _output_set = p_output_set;
n = inputSet.size(); _n = _input_set.size();
k = inputSet[0].size(); _k = _input_set[0].size();
lrScheduler = "None"; _lr_scheduler = "None";
decayConstant = 0; _decay_constant = 0;
dropRate = 0; _drop_rate = 0;
} }
MLPPANN::MLPPANN() { MLPPANN::MLPPANN() {
} }
MLPPANN::~MLPPANN() { MLPPANN::~MLPPANN() {
delete outputLayer; delete _output_layer;
} }
real_t MLPPANN::cost(std::vector<real_t> y_hat, std::vector<real_t> y) { real_t MLPPANN::cost(std::vector<real_t> y_hat, std::vector<real_t> y) {
@ -759,92 +762,92 @@ real_t MLPPANN::cost(std::vector<real_t> y_hat, std::vector<real_t> y) {
MLPPCost mlpp_cost; MLPPCost mlpp_cost;
real_t totalRegTerm = 0; real_t totalRegTerm = 0;
auto cost_function = outputLayer->cost_map[outputLayer->cost]; auto cost_function = _output_layer->cost_map[_output_layer->cost];
if (!network.empty()) { if (!_network.empty()) {
for (uint32_t i = 0; i < network.size() - 1; i++) { for (uint32_t i = 0; i < _network.size() - 1; i++) {
totalRegTerm += regularization.regTerm(network[i].weights, network[i].lambda, network[i].alpha, network[i].reg); totalRegTerm += regularization.regTerm(_network[i].weights, _network[i].lambda, _network[i].alpha, _network[i].reg);
} }
} }
return (mlpp_cost.*cost_function)(y_hat, y) + totalRegTerm + regularization.regTerm(outputLayer->weights, outputLayer->lambda, outputLayer->alpha, outputLayer->reg); return (mlpp_cost.*cost_function)(y_hat, y) + totalRegTerm + regularization.regTerm(_output_layer->weights, _output_layer->lambda, _output_layer->alpha, _output_layer->reg);
} }
void MLPPANN::forward_pass() { void MLPPANN::forward_pass() {
if (!network.empty()) { if (!_network.empty()) {
network[0].input = inputSet; _network[0].input = _input_set;
network[0].forwardPass(); _network[0].forwardPass();
for (uint32_t i = 1; i < network.size(); i++) { for (uint32_t i = 1; i < _network.size(); i++) {
network[i].input = network[i - 1].a; _network[i].input = _network[i - 1].a;
network[i].forwardPass(); _network[i].forwardPass();
} }
outputLayer->input = network[network.size() - 1].a; _output_layer->input = _network[_network.size() - 1].a;
} else { } else {
outputLayer->input = inputSet; _output_layer->input = _input_set;
} }
outputLayer->forwardPass(); _output_layer->forwardPass();
y_hat = outputLayer->a; _y_hat = _output_layer->a;
} }
void MLPPANN::update_parameters(std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations, std::vector<real_t> outputLayerUpdation, real_t learning_rate) { void MLPPANN::update_parameters(std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations, std::vector<real_t> output_layer_updation, real_t learning_rate) {
MLPPLinAlg alg; MLPPLinAlg alg;
outputLayer->weights = alg.subtraction(outputLayer->weights, outputLayerUpdation); _output_layer->weights = alg.subtraction(_output_layer->weights, output_layer_updation);
outputLayer->bias -= learning_rate * alg.sum_elements(outputLayer->delta) / n; _output_layer->bias -= learning_rate * alg.sum_elements(_output_layer->delta) / _n;
if (!network.empty()) { if (!_network.empty()) {
network[network.size() - 1].weights = alg.subtraction(network[network.size() - 1].weights, hiddenLayerUpdations[0]); _network[_network.size() - 1].weights = alg.subtraction(_network[_network.size() - 1].weights, hidden_layer_updations[0]);
network[network.size() - 1].bias = alg.subtractMatrixRows(network[network.size() - 1].bias, alg.scalarMultiply(learning_rate / n, network[network.size() - 1].delta)); _network[_network.size() - 1].bias = alg.subtractMatrixRows(_network[_network.size() - 1].bias, alg.scalarMultiply(learning_rate / _n, _network[_network.size() - 1].delta));
for (int i = network.size() - 2; i >= 0; i--) { for (int i = _network.size() - 2; i >= 0; i--) {
network[i].weights = alg.subtraction(network[i].weights, hiddenLayerUpdations[(network.size() - 2) - i + 1]); _network[i].weights = alg.subtraction(_network[i].weights, hidden_layer_updations[(_network.size() - 2) - i + 1]);
network[i].bias = alg.subtractMatrixRows(network[i].bias, alg.scalarMultiply(learning_rate / n, network[i].delta)); _network[i].bias = alg.subtractMatrixRows(_network[i].bias, alg.scalarMultiply(learning_rate / _n, _network[i].delta));
} }
} }
} }
std::tuple<std::vector<std::vector<std::vector<real_t>>>, std::vector<real_t>> MLPPANN::compute_gradients(std::vector<real_t> y_hat, std::vector<real_t> outputSet) { std::tuple<std::vector<std::vector<std::vector<real_t>>>, std::vector<real_t>> MLPPANN::compute_gradients(std::vector<real_t> y_hat, std::vector<real_t> _output_set) {
// std::cout << "BEGIN" << std::endl; // std::cout << "BEGIN" << std::endl;
MLPPCost mlpp_cost; MLPPCost mlpp_cost;
MLPPActivation avn; MLPPActivation avn;
MLPPLinAlg alg; MLPPLinAlg alg;
MLPPReg regularization; MLPPReg regularization;
std::vector<std::vector<std::vector<real_t>>> cumulativeHiddenLayerWGrad; // Tensor containing ALL hidden grads. std::vector<std::vector<std::vector<real_t>>> cumulative_hidden_layer_w_grad; // Tensor containing ALL hidden grads.
auto costDeriv = outputLayer->costDeriv_map[outputLayer->cost]; auto costDeriv = _output_layer->costDeriv_map[_output_layer->cost];
auto outputAvn = outputLayer->activation_map[outputLayer->activation]; auto outputAvn = _output_layer->activation_map[_output_layer->activation];
outputLayer->delta = alg.hadamard_product((mlpp_cost.*costDeriv)(y_hat, outputSet), (avn.*outputAvn)(outputLayer->z, 1)); _output_layer->delta = alg.hadamard_product((mlpp_cost.*costDeriv)(y_hat, _output_set), (avn.*outputAvn)(_output_layer->z, 1));
std::vector<real_t> outputWGrad = alg.mat_vec_mult(alg.transpose(outputLayer->input), outputLayer->delta); std::vector<real_t> output_w_grad = alg.mat_vec_mult(alg.transpose(_output_layer->input), _output_layer->delta);
outputWGrad = alg.addition(outputWGrad, regularization.regDerivTerm(outputLayer->weights, outputLayer->lambda, outputLayer->alpha, outputLayer->reg)); output_w_grad = alg.addition(output_w_grad, regularization.regDerivTerm(_output_layer->weights, _output_layer->lambda, _output_layer->alpha, _output_layer->reg));
if (!network.empty()) { if (!_network.empty()) {
auto hiddenLayerAvn = network[network.size() - 1].activation_map[network[network.size() - 1].activation]; auto hiddenLayerAvn = _network[_network.size() - 1].activation_map[_network[_network.size() - 1].activation];
network[network.size() - 1].delta = alg.hadamard_product(alg.outerProduct(outputLayer->delta, outputLayer->weights), (avn.*hiddenLayerAvn)(network[network.size() - 1].z, 1)); _network[_network.size() - 1].delta = alg.hadamard_product(alg.outerProduct(_output_layer->delta, _output_layer->weights), (avn.*hiddenLayerAvn)(_network[_network.size() - 1].z, 1));
std::vector<std::vector<real_t>> hiddenLayerWGrad = alg.matmult(alg.transpose(network[network.size() - 1].input), network[network.size() - 1].delta); std::vector<std::vector<real_t>> hiddenLayerWGrad = alg.matmult(alg.transpose(_network[_network.size() - 1].input), _network[_network.size() - 1].delta);
cumulativeHiddenLayerWGrad.push_back(alg.addition(hiddenLayerWGrad, regularization.regDerivTerm(network[network.size() - 1].weights, network[network.size() - 1].lambda, network[network.size() - 1].alpha, network[network.size() - 1].reg))); // Adding to our cumulative hidden layer grads. Maintain reg terms as well. cumulative_hidden_layer_w_grad.push_back(alg.addition(hiddenLayerWGrad, regularization.regDerivTerm(_network[_network.size() - 1].weights, _network[_network.size() - 1].lambda, _network[_network.size() - 1].alpha, _network[_network.size() - 1].reg))); // Adding to our cumulative hidden layer grads. Maintain reg terms as well.
for (int i = network.size() - 2; i >= 0; i--) { for (int i = _network.size() - 2; i >= 0; i--) {
hiddenLayerAvn = network[i].activation_map[network[i].activation]; hiddenLayerAvn = _network[i].activation_map[_network[i].activation];
network[i].delta = alg.hadamard_product(alg.matmult(network[i + 1].delta, alg.transpose(network[i + 1].weights)), (avn.*hiddenLayerAvn)(network[i].z, 1)); _network[i].delta = alg.hadamard_product(alg.matmult(_network[i + 1].delta, alg.transpose(_network[i + 1].weights)), (avn.*hiddenLayerAvn)(_network[i].z, 1));
hiddenLayerWGrad = alg.matmult(alg.transpose(network[i].input), network[i].delta); hiddenLayerWGrad = alg.matmult(alg.transpose(_network[i].input), _network[i].delta);
cumulativeHiddenLayerWGrad.push_back(alg.addition(hiddenLayerWGrad, regularization.regDerivTerm(network[i].weights, network[i].lambda, network[i].alpha, network[i].reg))); // Adding to our cumulative hidden layer grads. Maintain reg terms as well. cumulative_hidden_layer_w_grad.push_back(alg.addition(hiddenLayerWGrad, regularization.regDerivTerm(_network[i].weights, _network[i].lambda, _network[i].alpha, _network[i].reg))); // Adding to our cumulative hidden layer grads. Maintain reg terms as well.
} }
} }
return { cumulativeHiddenLayerWGrad, outputWGrad }; return { cumulative_hidden_layer_w_grad, output_w_grad };
} }
void MLPPANN::print_ui(int epoch, real_t cost_prev, std::vector<real_t> y_hat, std::vector<real_t> outputSet) { void MLPPANN::print_ui(int epoch, real_t cost_prev, std::vector<real_t> y_hat, std::vector<real_t> p_output_set) {
MLPPUtilities::CostInfo(epoch, cost_prev, cost(y_hat, outputSet)); MLPPUtilities::CostInfo(epoch, cost_prev, cost(y_hat, p_output_set));
std::cout << "Layer " << network.size() + 1 << ": " << std::endl; std::cout << "Layer " << _network.size() + 1 << ": " << std::endl;
MLPPUtilities::UI(outputLayer->weights, outputLayer->bias); MLPPUtilities::UI(_output_layer->weights, _output_layer->bias);
if (!network.empty()) { if (!_network.empty()) {
for (int i = network.size() - 1; i >= 0; i--) { for (int i = _network.size() - 1; i >= 0; i--) {
std::cout << "Layer " << i + 1 << ": " << std::endl; std::cout << "Layer " << i + 1 << ": " << std::endl;
MLPPUtilities::UI(network[i].weights, network[i].bias); MLPPUtilities::UI(_network[i].weights, _network[i].bias);
} }
} }
} }

30
mlpp/ann/ann.h
View File

@ -48,37 +48,37 @@ public:
void add_layer(int n_hidden, std::string activation, std::string weight_init = "Default", std::string reg = "None", real_t lambda = 0.5, real_t alpha = 0.5); void add_layer(int n_hidden, std::string activation, std::string weight_init = "Default", std::string reg = "None", real_t lambda = 0.5, real_t alpha = 0.5);
void add_output_layer(std::string activation, std::string loss, std::string weight_init = "Default", std::string reg = "None", real_t lambda = 0.5, real_t alpha = 0.5); void add_output_layer(std::string activation, std::string loss, std::string weight_init = "Default", std::string reg = "None", real_t lambda = 0.5, real_t alpha = 0.5);
MLPPANN(std::vector<std::vector<real_t>> inputSet, std::vector<real_t> outputSet); MLPPANN(std::vector<std::vector<real_t>> p_input_set, std::vector<real_t> p_output_set);
MLPPANN(); MLPPANN();
~MLPPANN(); ~MLPPANN();
protected: protected:
real_t apply_learning_rate_scheduler(real_t learningRate, real_t decayConstant, real_t epoch, real_t dropRate); real_t apply_learning_rate_scheduler(real_t learning_rate, real_t decay_constant, real_t epoch, real_t drop_rate);
real_t cost(std::vector<real_t> y_hat, std::vector<real_t> y); real_t cost(std::vector<real_t> y_hat, std::vector<real_t> y);
void forward_pass(); void forward_pass();
void update_parameters(std::vector<std::vector<std::vector<real_t>>> hiddenLayerUpdations, std::vector<real_t> outputLayerUpdation, real_t learning_rate); void update_parameters(std::vector<std::vector<std::vector<real_t>>> hidden_layer_updations, std::vector<real_t> output_layer_updation, real_t learning_rate);
std::tuple<std::vector<std::vector<std::vector<real_t>>>, std::vector<real_t>> compute_gradients(std::vector<real_t> y_hat, std::vector<real_t> outputSet); std::tuple<std::vector<std::vector<std::vector<real_t>>>, std::vector<real_t>> compute_gradients(std::vector<real_t> y_hat, std::vector<real_t> _output_set);
void print_ui(int epoch, real_t cost_prev, std::vector<real_t> y_hat, std::vector<real_t> outputSet); void print_ui(int epoch, real_t cost_prev, std::vector<real_t> y_hat, std::vector<real_t> p_output_set);
static void _bind_methods(); static void _bind_methods();
std::vector<std::vector<real_t>> inputSet; std::vector<std::vector<real_t>> _input_set;
std::vector<real_t> outputSet; std::vector<real_t> _output_set;
std::vector<real_t> y_hat; std::vector<real_t> _y_hat;
std::vector<MLPPOldHiddenLayer> network; std::vector<MLPPOldHiddenLayer> _network;
MLPPOldOutputLayer *outputLayer; MLPPOldOutputLayer *_output_layer;
int n; int _n;
int k; int _k;
std::string lrScheduler; std::string _lr_scheduler;
real_t decayConstant; real_t _decay_constant;
real_t dropRate; real_t _drop_rate;
}; };
#endif /* ANN_hpp */ #endif /* ANN_hpp */