MCPcopy Create free account

hub / github.com/KarelPeeters/Kyanite / functions

Functions1,306 in github.com/KarelPeeters/Kyanite

Function__nv_bfloat162 __lowhigh2highlow
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2258
Function__nv_bfloat162 __lows2bfloat162
defined(__CUDACC__) || defined(_NVHPC_CUDA) */
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2143
Function__nv_bfloat162 __shfl_down_sync
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2337
Function__nv_bfloat162 __shfl_sync
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2323
Function__nv_bfloat162 __shfl_up_sync
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2330
Function__nv_bfloat162 __shfl_xor_sync
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2344
Function__nv_bfloat162 atomicAdd
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:4305
Function__nv_bfloat162 h2ceil
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2125
Function__nv_bfloat162 h2cos
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3695
Function__nv_bfloat162 h2exp
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3737
Function__nv_bfloat162 h2exp10
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3791
Function__nv_bfloat162 h2floor
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2131
Function__nv_bfloat162 h2log
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3868
Function__nv_bfloat162 h2log10
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3910
Function__nv_bfloat162 h2rcp
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3945
Function__nv_bfloat162 h2rint
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2138
Function__nv_bfloat162 h2sin
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:3682
Function__nv_bfloat162 h2trunc
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:2119
Function__nv_bfloat162 make_bfloat162
CUDA vector-types compatible vector creation function (note returns __nv_bfloat162, not nv_bfloat162) */
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:1190
Function__nv_bfloat162_raw
* Conversion operator to \p __nv_bfloat162_raw */
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:792
Function__shfl
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:2357
Function__shfl_down
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:2371
Function__shfl_up
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:2364
Function__shfl_xor
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:2378
Function__short2half_rd
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1699
Function__short2half_ru
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1710
Function__short2half_rz
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1688
Function__uint2half_rd
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1767
Function__uint2half_ru
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1778
Function__uint2half_rz
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1756
Function__ull2half_rd
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1899
Function__ull2half_ru
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1910
Function__ull2half_rz
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1888
Function__ushort2half_rd
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1831
Function__ushort2half_ru
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1842
Function__ushort2half_rz
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:1820
Functionadd
()
kn-graph/src/onnx/typed_value.rs:275
Methodadd
(self, rhs: Self)
kn-graph/src/dtype.rs:729
Methodadd
(self, rhs: R)
kn-graph/src/shape.rs:424
Functionadd_broadcast
()
kn-cuda-eval/tests/root/graphs.rs:455
Functionaffine_add_twice
()
kn-cuda-eval/tests/root/graphs.rs:505
Functionaffine_multiple_channels
()
kn-cuda-eval/tests/root/graphs.rs:539
Functionaffine_padding
()
kn-cuda-eval/tests/root/graphs.rs:593
Functionaffine_single_div
()
kn-cuda-eval/tests/root/graphs.rs:526
Functionaffine_single_element
()
kn-cuda-eval/tests/root/graphs.rs:476
Methodalloc
(device: CudaDevice, len_bytes: usize)
kn-cuda-sys/src/wrapper/mem/device.rs:49
Methodalloc
TODO should we take a device param here?
kn-cuda-sys/src/wrapper/mem/pinned.rs:27
Methodalloc_simple
(device: CudaDevice, shape: Vec<usize>, dtype: DType)
kn-cuda-eval/src/device_tensor.rs:21
Methodas_str_name
String value of the enum field names used in the ProtoBuf definition. The values are not transformed in any way and thus are considered stable (if th
kn-graph/src/onnx/proto.rs:110
Methodas_string
(&self)
kn-cuda-sys/src/wrapper/status.rs:38
FunctionatomicAdd
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:3985
Functionatt_tiny
()
kn-cuda-eval/tests/root/onnx.rs:29
Functionbackend
()
kn-python/example.py:87
Methodbatched
(&self)
kn-graph/src/shape.rs:134
Functionbatched_mat_mul
(left: ArrayView3<T>, right: ArrayView3<T>)
kn-graph/src/cpu.rs:421
Methodbegin_capture
(&self)
kn-cuda-sys/src/wrapper/handle.rs:165
Methodbest
Returns the best available device. For now the algorithm used is very simple: pick the first cuda device if any are available otherwise use the CPU
kn-runtime/src/lib.rs:100
Functionbinary
()
kn-cuda-eval/tests/root/elementwise.rs:40
Functionbit_cast
()
kn-cuda-eval/tests/root/elementwise.rs:88
Methodbool
()
kn-graph/src/dtype.rs:188
Methodbranch
(tensor: &mut Tensor<T>, buffer: &[u8])
kn-cuda-eval/src/executor.rs:163
Methodbroadcast
(&self, new_shape: Vec<usize>)
kn-cuda-eval/src/offset_tensor.rs:66
Methodbuffer
(&self)
kn-cuda-sys/src/wrapper/mem/pool.rs:60
Functionc_array_string
(values: &[isize])
kn-cuda-eval/src/autokernel/common.rs:72
Functionc_nested_array_string
(values: &[Vec<isize>])
kn-cuda-eval/src/autokernel/common.rs:67
Functioncast_id
()
kn-cuda-eval/tests/root/opt.rs:125
Functionchain
()
kn-cuda-eval/tests/root/graphs.rs:734
Functionchained_scalar
()
kn-cuda-eval/tests/root/opt.rs:14
Functionchar
* Conversion operator to an implementation defined \p char data type. * Using round-toward-zero rounding mode. * * Detects signedness
kn-cuda-sys/doc_headers/cuda_include/cuda_fp16.hpp:452
Functionchar
* Conversion operator to an implementation defined \p char data type. * Using round-toward-zero rounding mode. * * Detects signedness
kn-cuda-sys/doc_headers/cuda_include/cuda_bf16.hpp:417
Functioncheck_cudnn
Check that the given graph produces the correct outputs as described by `check_data`, which typically comes from a `.bin` file next to the `.onnx` fil
kn-cuda-eval/src/tester.rs:14
Functioncheck_tensor_match_exact
( i: usize, expected_output: &Tensor<T>, output: &Tensor<T>, counts: &mut Counts, first_er
kn-cuda-eval/src/tester.rs:216
Functionclamp
()
kn-cuda-eval/tests/root/elementwise.rs:49
Methodclear
(&mut self)
kn-cuda-sys/src/wrapper/mem/pool.rs:36
Functioncollect_groups
(shape: &StridedShape)
kn-cuda-eval/src/shape.rs:372
Functioncompiled_with_cuda_support
Whether the crate was compiled with cuda support. This is independent of whether the current system actually has a cuda device available.
kn-runtime/src/lib.rs:59
Functioncomplex_multiply_new_axis
()
kn-cuda-eval/tests/root/graphs.rs:967
Functioncomplex_multiply_stride
()
kn-cuda-eval/tests/root/graphs.rs:962
Functionconcat
()
kn-cuda-eval/tests/root/graphs.rs:690
Functionconv_padding
()
kn-cuda-eval/tests/root/graphs.rs:566
Functioncopy
()
kn-cuda-eval/tests/root/graphs.rs:22
Methodcopy_from_host_staged
A (potentially) slower version of [Self::copy_simple_from_host] that works for any strides, by potentially copying to an intermediate stage on the dev
kn-cuda-eval/src/device_tensor.rs:130
Methodcopy_to_host_staged
A (potentially) slower version of [Self::copy_simple_to_host] that works for any strides, by potentially copying to an intermediate stage on the devic
kn-cuda-eval/src/device_tensor.rs:151
FunctioncuCabs
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:270
FunctioncuCabsf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:162
FunctioncuCadd
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:212
FunctioncuCaddf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:102
FunctioncuCdiv
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:247
FunctioncuCdivf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:137
FunctioncuCfma
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:334
FunctioncuCfmaf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:320
FunctioncuCmul
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:231
FunctioncuCmulf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:121
FunctioncuComplexDoubleToFloat
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:313
FunctioncuComplexFloatToDouble
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:307
FunctioncuConj
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:207
FunctioncuConjf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:98
FunctioncuCsub
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:219
FunctioncuCsubf
kn-cuda-sys/doc_headers/cuda_include/cuComplex.h:109
FunctioncuGetProcAddress_v2_ptsz
kn-cuda-sys/doc_headers/cuda_include/cuda.h:22661
← previousnext →701–800 of 1,306, ranked by callers