lisyarus/psemek
1
0
Fork 1
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Compare commits

base: lisyarus:c41e3874ca0edf1dbe4c62057dac41072c6b979f
Branches Tags
lisyarus:master
..
compare: lisyarus:614338a1b0b1f143cbff633597a105a7f3094794
Branches Tags
lisyarus:master

3 commits
c41e3874ca ... 614338a1b0

Author SHA1 Message Date
lisyarus
614338a1b0 Use smaller threshold for ml gradient tests
All checks were successful
Run tests / Run tests (push) Successful in 8m34s
Details
2026-09-12 20:07:33 +03:00
lisyarus
24b138b7e0 Add tests for ml activation derivatives 2026-09-12 20:05:39 +03:00
lisyarus
e378e24739 Fix softplus formula 2026-09-12 20:05:39 +03:00
2 changed files with 24 additions and 3 deletions
Show stats Expand all files Collapse all files

2
libs/ml/include/psemek/ml/neural_net/activation.hpp
View file

@ -36,7 +36,7 @@ namespace psemek::ml
case activation_type::softplus: case activation_type::softplus:
return (x < T{0}) return (x < T{0})
? std::log(T{1} + std::exp(x)) ? std::log(T{1} + std::exp(x))
: x + std::log(T{1} + std::exp(-1)); : x + std::log(T{1} + std::exp(-x));
case activation_type::squareplus: case activation_type::squareplus:
return (x < T{0}) return (x < T{0})
? T{2} / (std::sqrt(x * x + T{4}) - x) ? T{2} / (std::sqrt(x * x + T{4}) - x)

25
libs/ml/tests/neural_net/gradient.cpp
View file

@ -12,6 +12,27 @@ using namespace psemek::ml;
using namespace psemek::random; using namespace psemek::random;
using namespace psemek::math; using namespace psemek::math;
test_case(ml_neural__net_activation)
{
generator rng;
for (std::size_t i = 0; i < activation_type_values().size(); ++i)
{
auto a = static_cast<activation_type>(i);
for (std::size_t j = 0; j < 64; ++j)
{
double const eps = 1e-6;
double x = uniform(rng, -10.0, 10.0);
double value = activation(x, a);
double exact_derivative = activation_derivative(value, a);
double approx_derivative = (activation(x + eps, a) - value) / eps;
expect_close(exact_derivative, approx_derivative, 1e-6);
}
}
}
test_case(ml_neural__net_gradient) test_case(ml_neural__net_gradient)
{ {
generator rng; generator rng;
@ -54,7 +75,7 @@ test_case(ml_neural__net_gradient)
nn.weights()[i] = old; nn.weights()[i] = old;
double numeric_gradient = (v1 - v0) / 2.0 / eps; double numeric_gradient = (v1 - v0) / 2.0 / eps;
expect_close(numeric_gradient, learner.gradient()[i], 1e-4); expect_close(numeric_gradient, learner.gradient()[i], 1e-6);
} }
} }
} }
@ -101,7 +122,7 @@ test_case(ml_neural__net_arg__gradient)
input[i] = old; input[i] = old;
double numeric_gradient = (v1 - v0) / 2.0 / eps; double numeric_gradient = (v1 - v0) / 2.0 / eps;
expect_close(numeric_gradient, learner.arg_gradient()[i], 1e-4); expect_close(numeric_gradient, learner.arg_gradient()[i], 1e-6);
} }
} }
} }