MCPcopy Create free account
hub / github.com/0xShug0/audio.cpp / tensor_data

Function tensor_data

include/engine/framework/modules/weight_binding.h:31–33  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

29 float eps,
30 assets::TensorStorageType storage_type = assets::TensorStorageType::F32) {
31 const auto gamma = source.require_f32(prefix + ".weight", {channels});
32 const auto beta = source.require_f32(prefix + ".bias", {channels});
33 const auto mean = source.require_f32(prefix + ".running_mean", {channels});
34 const auto variance = source.require_f32(prefix + ".running_var", {channels});
35 std::vector<float> scale(static_cast<size_t>(channels)), bias(static_cast<size_t>(channels));
36 for (size_t i = 0; i < scale.size(); ++i) {

Callers 7

scale_channelsFunction · 0.85
linear_dataFunction · 0.85
norm_dataFunction · 0.85
conv1d_dataFunction · 0.85
conv2d_dataFunction · 0.85
conv_transpose1d_dataFunction · 0.85
layer_scale_dataFunction · 0.85

Calls 4

make_f32Method · 0.45
make_tensorMethod · 0.45
dataMethod · 0.45
sizeMethod · 0.45

Tested by

no test coverage detected