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

Function load_rms_norm

src/models/qwen3_tts/tokenizer_speech_decoder.cpp:329–335  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

327 weights.use_bias = false;
328 const auto data = source.require_tensor_as_shape(
329 prefix + ".weight",
330 storage_type,
331 {output_dim, input_dim, 1},
332 {output_dim, input_dim});
333 weights.weight = store.make_tensor(data.shape, data.type, data.bytes.data(), data.bytes.size());
334 return weights;
335}
336
337RmsNormWeights load_rms_norm(
338 const assets::TensorSource & source,

Callers 1

load_weightsFunction · 0.70

Calls 1

require_f32Method · 0.45

Tested by

no test coverage detected