Intel® oneAPI Deep Neural Network Developer Guide and Reference
A newer version of this document is available. Customers should click here to go to the newest version.
Visible to Intel only — GUID: GUID-9BD8D275-2E19-44C1-9A1F-40F0235715BB
Visible to Intel only — GUID: GUID-9BD8D275-2E19-44C1-9A1F-40F0235715BB
enum dnnl::memory::data_type
Overview
Data type specification. More…
#include <dnnl.hpp> enum data_type { undef = dnnl_data_type_undef, f8_e5m2 = dnnl_f8_e5m2, f8_e4m3 = dnnl_f8_e4m3, f16 = dnnl_f16, bf16 = dnnl_bf16, f32 = dnnl_f32, f64 = dnnl_f64, s32 = dnnl_s32, s8 = dnnl_s8, u8 = dnnl_u8, s4 = dnnl_s4, u4 = dnnl_u4, };
Detailed Documentation
Data type specification.
Enum Values
undef
Undefined data type (used for empty memory descriptors).
f8_e5m2
OFP8 standard 8-bit floating-point with a 5-bit exponent and a 2-bit mantissa.
f8_e4m3
OFP8 standard 8-bit floating-point with a 4-bit exponent and a 3-bit mantissa.
f16
16-bit/half-precision floating point.
bf16
non-standard 16-bit floating point with 7-bit mantissa.
f32
32-bit/single-precision floating point.
s32
32-bit signed integer.
s8
8-bit signed integer.
u8
8-bit unsigned integer.
s4
4-bit signed integer.
u4
4-bit unsigned integer.