hpt 0.1.3

High Performance Tensor (HPT) - A fast, efficient, and user-friendly tensor computation library for Rust
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121

#![allow(unused)]

use std::borrow::Borrow;

use crate::backends::cuda::utils::binary::binary_normal::binary_fn_precompiled;
use crate::backends::cuda::utils::binary::binary_normal::binary_fn_with_out_simd;
use crate::{tensor::Tensor, tensor_base::_Tensor, BoolVector};
use cudarc::driver::DeviceRepr;
use hpt_allocator::traits::{Allocator, AllocatorOutputRetrive};
use hpt_allocator::Cuda;
use hpt_common::error::base::TensorError;
use hpt_cudakernels::CMP;
use hpt_traits::{ops::cmp::TensorCmp, tensor::CommonBounds};
use hpt_types::cuda_types::scalar::Scalar;
use hpt_types::dtype::CudaType;
use hpt_types::{
    dtype::TypeCommon,
    into_vec::IntoVec,
    type_promote::{Cmp, SimdCmp},
};
impl<T, C, const DEVICE_ID: usize, Al> TensorCmp<T, C> for Tensor<T, Cuda, DEVICE_ID, Al>
where
    T: CommonBounds + DeviceRepr + Cmp<C> + CudaType,
    C: CommonBounds + DeviceRepr + CudaType,
    Scalar<T>: Cmp<Scalar<C>, Output = Scalar<bool>>,
    Al: Allocator,
    Al::Output: AllocatorOutputRetrive,
{
    type RHS = Tensor<C, Cuda, DEVICE_ID, Al>;
    type Output = Tensor<bool, Cuda, DEVICE_ID, Al>;

    fn tensor_neq<D>(&self, rhs: D) -> Result<Self::Output, TensorError>
    where
        D: Borrow<Self::RHS>,
    {
        let res = binary_fn_precompiled(
            self.inner.as_ref(),
            rhs.borrow().inner.as_ref(),
            "ne",
            &CMP,
            None::<_Tensor<bool, Cuda, DEVICE_ID, Al>>,
        )
        .unwrap();
        Ok(res.into())
    }

    fn tensor_eq<D>(&self, rhs: D) -> Result<Self::Output, TensorError>
    where
        D: Borrow<Self::RHS>,
    {
        let res = binary_fn_precompiled(
            self.inner.as_ref(),
            rhs.borrow().inner.as_ref(),
            "eq",
            &CMP,
            None::<_Tensor<bool, Cuda, DEVICE_ID, Al>>,
        )
        .unwrap();
        Ok(res.into())
    }

    fn tensor_lt<D>(&self, rhs: D) -> Result<Self::Output, TensorError>
    where
        D: Borrow<Self::RHS>,
    {
        let res = binary_fn_precompiled(
            self.inner.as_ref(),
            rhs.borrow().inner.as_ref(),
            "lt",
            &CMP,
            None::<_Tensor<bool, Cuda, DEVICE_ID, Al>>,
        )
        .unwrap();
        Ok(res.into())
    }

    fn tensor_gt<D>(&self, rhs: D) -> Result<Self::Output, TensorError>
    where
        D: Borrow<Self::RHS>,
    {
        let res = binary_fn_precompiled(
            self.inner.as_ref(),
            rhs.borrow().inner.as_ref(),
            "gt",
            &CMP,
            None::<_Tensor<bool, Cuda, DEVICE_ID, Al>>,
        )
        .unwrap();
        Ok(res.into())
    }

    fn tensor_le<D>(&self, rhs: D) -> Result<Self::Output, TensorError>
    where
        D: Borrow<Self::RHS>,
    {
        let res = binary_fn_precompiled(
            self.inner.as_ref(),
            rhs.borrow().inner.as_ref(),
            "le",
            &CMP,
            None::<_Tensor<bool, Cuda, DEVICE_ID, Al>>,
        )
        .unwrap();
        Ok(res.into())
    }

    fn tensor_ge<D>(&self, rhs: D) -> Result<Self::Output, TensorError>
    where
        D: Borrow<Self::RHS>,
    {
        let res = binary_fn_precompiled(
            self.inner.as_ref(),
            rhs.borrow().inner.as_ref(),
            "ge",
            &CMP,
            None::<_Tensor<bool, Cuda, DEVICE_ID, Al>>,
        )
        .unwrap();
        Ok(res.into())
    }
}