tensor.reduce_log_sum_exp
fn reduce_log_sum_exp(self: @Tensor<T>, axis: usize, keepdims: bool) -> Tensor<T>;
Computes the log sum of the exponentials of the input tensor's elements along the provided axes.
use core::array::{ArrayTrait, SpanTrait};
use orion::operators::tensor::{TensorTrait, Tensor};
use orion::operators::tensor::FP32x32Tensor;
use orion::numbers::{FixedTrait, FP32x32};
fn reduce_log_sum_exp() -> Tensor<FP32x32> {
let mut shape = ArrayTrait::<usize>::new();
shape.append(3);
shape.append(2);
shape.append(2);
let mut data = ArrayTrait::new();
data.append(FP32x32 { mag: 4294967296, sign: false });
data.append(FP32x32 { mag: 8589934592, sign: false });
data.append(FP32x32 { mag: 12884901888, sign: false });
data.append(FP32x32 { mag: 17179869184, sign: false });
data.append(FP32x32 { mag: 21474836480, sign: false });
data.append(FP32x32 { mag: 25769803776, sign: false });
data.append(FP32x32 { mag: 30064771072, sign: false });
data.append(FP32x32 { mag: 34359738368, sign: false });
data.append(FP32x32 { mag: 38654705664, sign: false });
data.append(FP32x32 { mag: 42949672960, sign: false });
data.append(FP32x32 { mag: 47244640256, sign: false });
data.append(FP32x32 { mag: 51539607552, sign: false });
TensorTrait::new(shape.span(), data.span())
let tensor = TensorTrait::<FP32x32>::new(shape.span(), data.span());
return tensor.reduce_log_sum_exp(axis: 2, keepdims: false);
}
>>> [[9215828, 16323477, 20115004], [22716772, 24699744, 26302432]]