float16

There are 0 repository under float16 topic.

NVIDIA / OpenSeq2Seq
Toolkit for efficient experimentation with Speech Recognition, Text2Speech and NLP
deep-learning float16 language-model mixed-precision multi-gpu multi-node neural-machine-translation seq2seq sequence-to-sequence speech-recognition speech-synthesis speech-to-text tensorflow text-to-speech
Language:Python 1537
SimSIMD
ashvardanian / SimSIMD
Up to 200x Faster Inner Products and Vector Similarity — for Python, JavaScript, Rust, and C, supporting f64, f32, f16 real & complex, i8, and binary vectors using SIMD for both x86 AVX2 & AVX-512 and Arm NEON & SVE 📐
arm-neon arm-sve assembly avx2 avx512 blas blas-libraries distance-calculation distance-measures float16 information-retrieval metrics neon numpy scipy simd simd-instructions similarity-measures similarity-search vector-search
Language:C 733
starkat99 / half-rs
Half-precision floating point types f16 and bf16 for Rust.
bfloat16 binary16 crates f16 float16 floating-point ieee754 rust rust-embedded
Language:Rust 217
petamoriken / float16
Stage 3 IEEE 754 half-precision floating-point ponyfill
binary16 float16 float16array fp16 half-precision ieee754 javascript typescript
Language:JavaScript 94
x448 / float16
float16 provides IEEE 754 half-precision format (binary16) with correct conversions to/from float32
binary16 float16 ieee754 floating-point half-precision go golang
Language:Go 61
andreped / GradientAccumulator
:dart: Accumulated Gradients for TensorFlow 2
tensorflow accumulated-gradients tensorflow2 tf2 batch-size memory-constraints gradient-accumulation distributed-training float16 mixed-precision deep-learning keras adaptive-gradient-clipping gpu tpu huggingface accumulated-batch-normalization multi-gpu hacktoberfest
Language:Python 48
hma02 / cublasHgemm-P100
Code for testing the native float16 matrix multiplication performance on Tesla P100 and V100 GPU based on cublasHgemm
gpu precision float16 half-precision p100 cublas gemm v100
Language:Cuda 35
artyom-beilis / float16
half float library for C and for z80
z80 z80asm c half-precision float16 zx-spectrum
Language:C 32
fengwang / float16_t
CPP20 implementation of a 16-bit floating-point type mimicking most of the IEEE 754 behavior. Single file and header-only.
half-precision float16 fp16 floating-point cpp20 header-only
Language:C++ 28
aredden / torch-cublas-hgemm
PyTorch half precision gemm lib w/ fused optional bias + optional relu/gelu
cuda float16 gemm pytorch
Language:Cuda 18
RobTillaart / float16
Arduino library to implement float16 data type
arduino float16
Language:C++ 15
d4l3k / go-bfloat16
Bfloat16 conversion utilities for Go/Golang
bf16 bfloat16 binary16 float16 go golang
Language:Go 3
dansav / clip-sharp
A basic library for using OpenAI:s CLIP model in C#
clip csharp float16 onnx
Language:C# 3
djzenma / TFLite-Applications-And-Models
TFLite applications: Optimized .tflite models (i.e. lightweight and low latency) and code to run directly on your Microcontroller!
tensorflow tensorflow-lite neural-network microcontroller embedded-systems tflite tflite-models benchmark analysis quantization dynamic-range float16 tinyml tflite-applications
3
junkwarrior87 / half-float
Utility for converting 16-bit floats
float float16 floating-point half-float
Language:TypeScript 2
stdlib-js / constants-float16
Half-precision floating-point mathematical constants.
nodejs javascript stdlib node node-js mathematics math standard library lib constants const namespace float16 single half-precision floating-point float ieee754
Language:JavaScript 2
stdlib-js / constants-float16-eps
Difference between one and the smallest value greater than one that can be represented as a half-precision floating-point number.
nodejs javascript stdlib node node-js constant const mathematics math float real half floating-point float16 16bit 16-bit ieee754 epsilon eps number
Language:JavaScript 2
stdlib-js / constants-float16-exponent-bias
The bias of a half-precision floating-point number's exponent.
nodejs javascript stdlib node node-js constant const mathematics math half precision floating-point float float16 16bit 16-bit ieee754 bias exponent
Language:JavaScript 2
stdlib-js / constants-float16-max
Maximum half-precision floating-point number.
nodejs javascript stdlib node node-js constant const mathematics math max maximum floating-point float16 float 16bit 16-bit ieee754 half precision largest
Language:JavaScript 2
stdlib-js / constants-float16-max-safe-integer
Maximum safe half-precision floating-point integer.
nodejs javascript stdlib node node-js constant const max maximum safe integer half floating point floating-point float float16 16-bit 16bit ieee754
Language:JavaScript 2
stdlib-js / constants-float16-min-safe-integer
Minimum safe half-precision floating-point integer.
nodejs javascript stdlib node node-js constant const min minimum safe integer half floating point floating-point float float16 16bit 16-bit ieee754
Language:JavaScript 2
stdlib-js / constants-float16-pinf
Half-precision floating-point positive infinity.
nodejs javascript stdlib node node-js constant const mathematics math pinf positive infinity float16 16bit 16-bit inf half precision floating-point float
Language:JavaScript 2
stdlib-js / constants-float16-smallest-subnormal
Smallest positive half-precision floating-point subnormal number.
nodejs javascript stdlib node node-js constant const mathematics math float float16 16bit 16-bit half positive nonzero min minimum smallest floating-point
Language:JavaScript 2
stdlib-js / constants-float16-sqrt-eps
Square root of half-precision floating-point epsilon.
nodejs javascript stdlib node node-js constant const mathematics math float real half half-precision floating-point float16 16bit 16-bit ieee754 epsilon eps
Language:JavaScript 2
TheLazyTomcat / lib.Float16Utils
Main purpose of this library is to provide functions for conversion to and from half precision (16bit) floating point numbers. It also provides functions for basic arithmetic and comparison of half floats.
delphi float16 floating-point fpc lazarus object-pascal pascal
Language:Pascal 2
tirfil / vhdfp16
FP16 Half precision floating point (IEEE754 2008) adder + multiplier
float16 ieee754 vhdl
Language:VHDL 2
stdlib-js / constants-float16-cbrt-eps
Cube root of half-precision floating-point epsilon.
nodejs javascript stdlib node node-js constant const mathematics math float real half half-precision floating-point float16 16bit 16-bit ieee754 epsilon eps
Language:JavaScript 1
stdlib-js / constants-float16-ninf
Half-precision floating-point negative infinity.
nodejs javascript stdlib node node-js constant const mathematics math ninf negative infinity float16 16bit 16-bit inf half precision floating-point float
Language:JavaScript 1
stdlib-js / constants-float16-smallest-normal
Smallest positive normalized half-precision floating-point number.
nodejs javascript stdlib node node-js constant const mathematics math float float16 16bit 16-bit half positive nonzero min minimum smallest floating-point
Language:JavaScript 1
joeltg / fp16
Half-precision 16-bit floating point numbers
typescript float16 half-precision floating-point
Language:JavaScript 0
yowidin / fast-half-float
Fast Half precision Floating point operations for C++
float float16 floating-point half-precision cpp20
Language:C++

float16

NVIDIA / OpenSeq2Seq

ashvardanian / SimSIMD

starkat99 / half-rs

petamoriken / float16

x448 / float16

andreped / GradientAccumulator

hma02 / cublasHgemm-P100

artyom-beilis / float16

fengwang / float16_t

aredden / torch-cublas-hgemm

RobTillaart / float16

d4l3k / go-bfloat16

dansav / clip-sharp

djzenma / TFLite-Applications-And-Models

junkwarrior87 / half-float

stdlib-js / constants-float16

stdlib-js / constants-float16-eps

stdlib-js / constants-float16-exponent-bias

stdlib-js / constants-float16-max

stdlib-js / constants-float16-max-safe-integer

stdlib-js / constants-float16-min-safe-integer

stdlib-js / constants-float16-pinf

stdlib-js / constants-float16-smallest-subnormal

stdlib-js / constants-float16-sqrt-eps

TheLazyTomcat / lib.Float16Utils

tirfil / vhdfp16

stdlib-js / constants-float16-cbrt-eps

stdlib-js / constants-float16-ninf

stdlib-js / constants-float16-smallest-normal

joeltg / fp16

yowidin / fast-half-float