1ca28fa995
CMSIS-DSP: Fix signed/unsigned integer data type issue
5 years ago
43f79ca92b
CMSIS-DSP: Improvements to Helium implementation of RFFT Q15 and Q31.
5 years ago
49a4c065b8
docs: fix simple typo, accumlate -> accumulate
...
There is a small typo in CMSIS/DSP/Source/FilteringFunctions/arm_conv_fast_opt_q15.c, CMSIS/DSP/Source/FilteringFunctions/arm_conv_opt_q15.c, CMSIS/DSP/Source/FilteringFunctions/arm_conv_opt_q7.c, CMSIS/DSP/Source/FilteringFunctions/arm_conv_partial_fast_opt_q15.c, CMSIS/DSP/Source/FilteringFunctions/arm_conv_partial_opt_q15.c, CMSIS/DSP/Source/FilteringFunctions/arm_conv_partial_opt_q7.c, CMSIS/DSP/Source/FilteringFunctions/arm_correlate_fast_opt_q15.c, CMSIS/DSP/Source/FilteringFunctions/arm_correlate_opt_q15.c, CMSIS/DSP/Source/FilteringFunctions/arm_correlate_opt_q7.c.
Should read `accumulate` rather than `accumlate`.
5 years ago
a888f7205b
CMSIS-DSP: Some minor corrections.
...
Documentation and #define naming.
5 years ago
c080770065
Add missing if clause
...
to not build float16 code if the datatype is disabled
5 years ago
8fff9ebe29
CMSIS-DSP: New MVE implementations of the complex dot product and vector product.
5 years ago
c4283d209f
CMSIS-DSP: Added new MVE implementation for FIR Q15 and Q7.
5 years ago
a108d6763e
CMSIS-DSP: Added new MVE implementation of FIR Q31
5 years ago
68b219bb1f
CMSIS-DSP: New MVE implementation of the FIR F16
5 years ago
364fdb9a28
CMSIS-DSP: Cleaning to MVE code for fir f32
5 years ago
6f229b4f65
CMSIS-DSP: New MVE implementation of the FIR f32
5 years ago
af027f97c5
CMSIS-DSP: Biquad improvement
...
Improvement to formatting script.
5 years ago
212e9cb805
CMSIS-DSP: Added Cholesky decomposition
...
LL^t and LDL^t variant.
Added foward and backward substitution to solve linear systems
with triangular matrix.
5 years ago
5409d06050
CMSIS-DSP: Fix FIR document
5 years ago
a7fd426546
CMSIS-DSP: Removed patch for armclang issue on M55
...
Issue will be corrected in 6.15
5 years ago
11ead8bc19
Correct DSP Documents: Correct the name of people.
5 years ago
0bcb1384f2
CMSIS-DSP: More consistent use of the flag ARM_MATH_AUTOVECTORIZE
5 years ago
7accd563b9
CMSIS-DSP: Corrected script for html generation
5 years ago
078c0b282d
CMSIS-DSP: Improved f16 scalar code.
5 years ago
4b4d5322c8
CMSIS-DSP: Improvement to f16 helium code for linear SVM.
...
Improved tests for weighted sum.
5 years ago
29a69f56d0
CMSIS-DSP: Improved some f16 helium codes.
5 years ago
8ba4243dfa
Correct DSP document: Correct the size of the referenced sine table.
5 years ago
3d1e329f33
CMSIS-DSP: Improved build with MDK
...
Compilation flags to control code size can now be used with MDK build
and not only with the cmake one.
5 years ago
8d2fca6efe
CMSIS-DSP: Removed compilation warning.
5 years ago
2b6d208557
CMSIS-DSP: Corrected issue in rfft f16
5 years ago
68e5b742ba
CMSIS-DSP: Some improvement to a few f16 functions
5 years ago
3752e622b8
CMSIS-DSP: Corrected issue in scalar arm_rfft_fast_f16
5 years ago
1b34e464ba
CMSIS-DSP: Added new f16 tests
5 years ago
69d56b1a31
CMSIS-DSP: f16 versions of classical ML functions
5 years ago
f51b3c3317
CMSIS-DSP: Added f16 versions of the distance functions
5 years ago
55c9be8af0
CMSIS-DSP: Added additional f16 statistics functions
...
and the required f16 fast math functions.
5 years ago
534c34f883
CMSIS-DSP: Added f16 support functions
5 years ago
51be1901ce
CMSIS-DSP: Added f16 versions of statistics functions.
5 years ago
71218873eb
CMSIS-DSP: Added f16 versions of linear and bilinear interpolations
5 years ago
d2d691cc23
CMSIS-DSP: Added f16 matrix functions
5 years ago
7d79ffa51f
CMSIS-DSP: Added arm_correlate_f16
...
Corrected PythonWrapper issues due to new f16 functions
5 years ago
f0750e92c2
CMSIS-DSP: Added f16 version of biquads.
...
Corrected few doxygen issues.
5 years ago
7c55ae80f7
CMSIS-DSP: Added arm_fir_f16
...
Improved f16 comlex dot product
Correction to compile flags to FFT tables for MVE version.
5 years ago
ebf9104c4e
CMSIS-DSP: Added complex math f16
5 years ago
8b465544a1
CMSIS-DSP: Added arm_rfft_fast_f16
5 years ago
9e4b2a9358
CMSIS-DSP: Use __PKHBT intrinsics for FFTs
...
This results in about a few percent faster FFT when using GCC.
Change-Id: I9620370aa5098bb10e04d84a7fed2167e7734aeb
5 years ago
41f2015387
CMSIS-DSP: Re-organization of arm_math.h
...
arm_math.h splitted into several headers.
Interpolation functions moved from arm_math.h to a separate folder.
5 years ago
93196c0fd7
CMSIS-DSP: Corrected build issue with Neon.
5 years ago
566550e15d
CMSIS-DSP: doxygen corrections
6 years ago
3359d706cf
CMSIS-DSP: Corrected build issue with Python wrapper.
...
Corrected bug in arm_mat_vec_mult_q15.
6 years ago
a20e215bfc
CMSIS-DSP: Added new matrix functions and their MVE implementation
...
arm_mat_cmplx_trans_f32.c
arm_mat_cmplx_trans_q15.c
arm_mat_cmplx_trans_q31.c
6 years ago
8268b079d5
CMSIS-DSP: Added new matrix functions and MVE version
...
arm_mat_mult_q7.c
arm_mat_trans_q7.c
arm_mat_vec_mult_f32.c
arm_mat_vec_mult_q15.c
arm_mat_vec_mult_q31.c
arm_mat_vec_mult_q7.c
6 years ago
c0d3b61f37
CMSIS-DSP: Changed cmake required minimum version
6 years ago
be40746c5e
CMSIS-DSP: Improvement to test framework
...
Added support for A32.
Still some issues with cycle model.
6 years ago
23d216b3bb
CMSIS-NN: Fix warnings
...
1. Doxygen warnings are fixed
2. Sign conversion warnings(-Wsign-conversion) and -Wextra warnings
are fixed for TFLu compliant MVE APIs
Change-Id: I22ee8cd4ff4c94c92d9659504fc4c24d2b5ae441
6 years ago
Yin-Cheng-888
Christophe Favergeon
Tim Gates
Lennart
Kishinami
Richard Allen
Felix Johnny