numr 0.5.1

High-performance numerical computing with multi-backend GPU acceleration (CPU/CUDA/WebGPU)
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
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136

//! CUDA implementation of 2:4 structured sparsity operations.

use crate::dtype::DType;
use crate::error::{Error, Result};
use crate::ops::traits::Sparse24Ops;
use crate::runtime::cuda::kernels::{
    launch_sparse_24_decompress, launch_sparse_24_matmul, launch_sparse_24_prune,
};
use crate::runtime::cuda::{CudaClient, CudaRuntime};
use crate::runtime::ensure_contiguous;
use crate::sparse::structured::{Sparse24Tensor, meta_cols_for_k};
use crate::tensor::Tensor;

impl Sparse24Ops<CudaRuntime> for CudaClient {
    fn prune_to_24(&self, dense: &Tensor<CudaRuntime>) -> Result<Sparse24Tensor<CudaRuntime>> {
        if dense.ndim() != 2 {
            return Err(Error::InvalidArgument {
                arg: "dense",
                reason: format!("Expected 2D tensor, got {}D", dense.ndim()),
            });
        }

        let m = dense.shape()[0];
        let k = dense.shape()[1];

        if !k.is_multiple_of(4) {
            return Err(Error::InvalidArgument {
                arg: "dense",
                reason: format!("K dimension ({k}) must be divisible by 4 for 2:4 sparsity"),
            });
        }

        let dtype = dense.dtype();
        let dense_contig = ensure_contiguous(dense);
        let half_k = k / 2;
        let mc = meta_cols_for_k(k);

        let compressed = Tensor::<CudaRuntime>::empty(&[m, half_k], dtype, &self.device);
        // Metadata must be zeroed before kernel's atomic OR operations
        let metadata = Tensor::<CudaRuntime>::zeros(&[m, mc], DType::U32, &self.device);

        unsafe {
            launch_sparse_24_prune(
                &self.context,
                &self.stream,
                self.device.index,
                dtype,
                dense_contig.ptr(),
                compressed.ptr(),
                metadata.ptr(),
                m,
                k,
            )?;
        }

        Sparse24Tensor::new(compressed, metadata, [m, k])
    }

    fn sparse_24_to_dense(
        &self,
        sparse: &Sparse24Tensor<CudaRuntime>,
    ) -> Result<Tensor<CudaRuntime>> {
        let [m, k] = sparse.shape();
        let dtype = sparse.dtype();

        let dense = Tensor::<CudaRuntime>::empty(&[m, k], dtype, &self.device);

        let vals = ensure_contiguous(sparse.compressed_values());
        let meta = ensure_contiguous(sparse.metadata());

        unsafe {
            launch_sparse_24_decompress(
                &self.context,
                &self.stream,
                self.device.index,
                dtype,
                vals.ptr(),
                meta.ptr(),
                dense.ptr(),
                m,
                k,
            )?;
        }

        Ok(dense)
    }

    fn sparse_24_matmul(
        &self,
        input: &Tensor<CudaRuntime>,
        weight: &Sparse24Tensor<CudaRuntime>,
    ) -> Result<Tensor<CudaRuntime>> {
        if input.ndim() != 2 {
            return Err(Error::InvalidArgument {
                arg: "input",
                reason: format!("Expected 2D tensor, got {}D", input.ndim()),
            });
        }

        let n = input.shape()[0];
        let input_k = input.shape()[1];
        let [m, weight_k] = weight.shape();

        if input_k != weight_k {
            return Err(Error::ShapeMismatch {
                expected: vec![n, weight_k],
                got: vec![n, input_k],
            });
        }

        let dtype = input.dtype();
        let input_contig = ensure_contiguous(input);
        let vals = ensure_contiguous(weight.compressed_values());
        let meta = ensure_contiguous(weight.metadata());

        let output = Tensor::<CudaRuntime>::empty(&[n, m], dtype, &self.device);

        unsafe {
            launch_sparse_24_matmul(
                &self.context,
                &self.stream,
                self.device.index,
                dtype,
                input_contig.ptr(),
                vals.ptr(),
                meta.ptr(),
                output.ptr(),
                n,
                m,
                weight_k,
            )?;
        }

        Ok(output)
    }
}