Method build

src/framework/modules/optimizations/fast_projection_modules.cpp:55–91

}

core::TensorValue FastPackedProjection4Module::build(
    core::ModuleBuildContext & ctx,
    const core::TensorValue & input,
    const LinearWeights & weights) const {
    if (ctx.ggml == nullptr) {
        throw std::runtime_error("ModuleBuildContext.ggml is null");
    }
    if (ctx.backend_type != core::BackendType::Cuda) {
        throw std::runtime_error("FastPackedProjection4Module is CUDA-only");
    }

    core::validate_rank_between(input, 1, core::kMaxTensorRank, "input");
    core::validate_last_dim(input, config_.in_features, "input");
    core::validate_shape(
        weights.weight,
        core::TensorShape::from_dims({config_.out_features, config_.in_features}),
        "weight");
    if (weights.bias.has_value()) {
        throw std::runtime_error("FastPackedProjection4Module does not support bias");
    }

    const auto contiguous_input = core::ensure_backend_addressable_layout(ctx, input);
    const auto matrix_input_shape = flatten_to_matrix_shape(contiguous_input.shape);
    const auto matrix_input = core::reshape_tensor(ctx, contiguous_input, matrix_input_shape);

    ggml_tensor * projected_raw =
        ggml_mul_mat_pack4(ctx.ggml, weights.weight.tensor, matrix_input.tensor);
    if (config_.precision != GGML_PREC_DEFAULT) {
        ggml_mul_mat_set_prec(projected_raw, config_.precision);
    }

    auto projected = core::wrap_tensor(
        projected_raw,
        core::TensorShape::from_dims({matrix_input_shape.at(0), config_.out_features}),
        GGML_TYPE_F32);
    return core::reshape_tensor(ctx, projected, input.shape.with_last_dim(config_.out_features));
}

const core::ModuleSchema & FastPackedProjection4Module::static_schema() noexcept {
    return kFastPackedProjection4Schema;