441f779647
Added a cross reference to compute graph README.
3 years ago
18d3504390
Corrected issue #85
...
Add compilation tests for MIN/MAX macro definitions.
3 years ago
d1bf6fe5d1
Update compute graph nodes to remove overriding issue.
3 years ago
1378fa94f5
Improved compute graph documentation
...
More details about the new dynamic mode.
3 years ago
fc38d7d21f
Some corrections to the compute graph README.
3 years ago
fab8a1e6a8
Correction to python wrapper build so that the result can be used on google colab
3 years ago
d1a2633be0
Updated PythonWrapper version and README
3 years ago
31d9d2ef4a
Missing file in pack. Small correction for compute graph.
3 years ago
de04777329
Corrected path in pack generation script
3 years ago
66ea68b096
Added a dynamic mode to the compute graph.
...
The compute graph hierarchy (Python and C++) has been modified.
It will break the code using the CG and the include and import will
have to be modified.
New function prepareForRunning is needed in CG nodes even if not used in
static mode. Without this function, code using compute graph won't build.
3 years ago
82559adce2
Added DTW to the PythonWrapper API
3 years ago
b46a2f86b5
First version of the dynamic time warping algorithm.
...
F32 only. Some windows suppported.
3 years ago
5bee006eaa
Added some documentation for the compute graph.
...
Some details about the inner working of the schedule generation algorithm.
3 years ago
a8f4056f79
use 32-bit bitreversal loop index ( #81 )
...
With GCC 8.3.1:
arm_bitreversal_16 - saves 1 instruction
arm_bitreversal_64 - saves 5 instructions
3 years ago
bd4b47a205
fix: typo from algorthms to algorithms ( #84 )
3 years ago
f757add3bb
Added window functions for use with Fourier transforms.
...
Regular functions for use with tone detection and noise estimation.
Flat-top windows for estimation of tone amplitudes.
Parameters of the windows are given.
Kaiser window missing (Bessel function not available in the library).
No f16 implementation since it is not accurate enough. It is better
to compute a f32 window and convert to f16 after.
3 years ago
8ec25f7a9e
Add braces around MAX macro ( #77 )
...
The current implementation leads to wrong calculation of output lengths.
3 years ago
1e84880535
Converted atan2_coefs_q15 to `signed int` constants ( #75 )
...
Removes conversion warning with `-pedantic`
3 years ago
258f31c5e6
Corrected issue #72
...
arm_sqrt_f32 optimizations for clang and gcc.
3 years ago
f91c1f3c24
Better test automation with cmsis build tools
...
Added a python script for automation of tests with cmsis build tools.
Correction to some tests
Float promotion issues in some f16 code.
3 years ago
43aa2a9e7f
Corrected pack generation script.
...
Missing header files.
3 years ago
db4253f0ee
New f16 files were not testing for ARM_FLOAT16_SUPPORTED
3 years ago
989370fe55
Corrected some python wrapper tests.
...
The deprecated rfft f32 has been removed from the wrapper
(rfft fast f32 is still included).
3 years ago
4d090bed42
Corrected Python wrapper for barycenter.
3 years ago
7a27992cdc
Apply -flax-vector-conversions flag to GCC ( #65 )
...
Some CMSIS-DSP code does not explicitly convert between vectors of
different numbers and types of elements. While some other compilers
default to implicitly converting vectors, gcc does not do this unless
explicitly enabled.
Signed-off-by: Lingkai Dong <lingkai.dong@arm.com>
Signed-off-by: Lingkai Dong <lingkai.dong@arm.com>
3 years ago
25c3fa3c02
Corrected compilation warnings.
3 years ago
9c425b04b9
Corrected doxygen generation script
3 years ago
633ab5913e
Add Conversion functions from and to 64 bit floating point
...
Cleaned f64 conversion functions.
3 years ago
60e1e88f45
doc(args): make sure function declarations have the same arguments names ( #67 )
...
Co-authored-by: Koen <koen@gsf.engineering>
3 years ago
8d5aae1e01
Add root CMakeLists.txt to better support FetchContent ( #64 )
...
CMake's `FetchContent_MakeAvailable()` treats the fetched repository as
a CMake project and automatically calls `subdirectory()` on it, if a
`CMakeLists.txt` exists in the root directory.
Signed-off-by: Lingkai Dong <lingkai.dong@arm.com>
Signed-off-by: Lingkai Dong <lingkai.dong@arm.com>
3 years ago
91ef795234
Corrected issue #66
...
Saturation problem in arm_absmax_no_idx_q(7|15|31)
Difference of behavior for absmax_q(7|15|31) between reference and code:
reference is first max found. Code was last max found.
3 years ago
3d1a7f7ff4
Improvement to the Python wrapper
...
Corrected issues with arm_fir_decimate and arm_fir_interpolate
Corrected issues with real FFTs in the wrapper
Added a customization option for the FIFO class in compute graph.
Added Python tests for the corrected functions.
3 years ago
0705c67568
Corrected issue #59
...
Out of bound access for arm_correlate MVE implementations.
3 years ago
54be9aeaee
Corrected issue #61
...
Table configuration issue with RFFT Fast F32 and F16
The define use to include or exclude the bit reverse table is different
for scalar and MVE versions. The test was not covering both cases.
3 years ago
68ddaa577d
Corrected issue #62
...
MVE tables were not const qualified.
3 years ago
bfda7097d5
Improvements to the compute graph documentation.
...
Some compute graph nodes have been moved out of the GenericNodes.h header
into their specific headers.
3 years ago
9eb4e92e74
Corrected issue #57
...
Problem with arm_scale_q15 and q31.
Reorganized a little the cmsis build scripts for the test framework.
3 years ago
b9b58d044e
Small change to a PythonWrapper test.
3 years ago
302897a523
Update to the release notes for the Python wrapper
3 years ago
f2f099f886
ComputeGraph improvements.
...
More customization options
3 years ago
568e45afc6
Improvements to CG
...
Small bug fix to make it easier to use several scheduler.
Improvement to declare new arguments for a node.
3 years ago
5a99a5c5ad
Improvement to compute graph
...
More customization options
New (experimental) scheduling algorithm to improve latencies.
3 years ago
8ed2ef7a7e
Corrected doxygen issue.
3 years ago
3d5fb83720
Add some f64 Neon optimisations for aarch64
...
Some documentation improvements.
3 years ago
633b528421
Corrected issue #49
...
Incorrect out of bounds value for linear interpolation (float).
3 years ago
f74f666fb0
Added more customization hooks to compute graph.
3 years ago
b8177102d9
Added support for CMSIS Event Recorder to the Compute Graph.
...
Added also more customization options for the compute graph.
3 years ago
142dcec5c3
Correction in python script for compute graph
...
Buffer numbering issue for memory optimzation and buffer sharing
3 years ago
06c439f7e7
Added documentation for cyclo static scheduling
3 years ago
30be135940
Modified C++ implementation of compute graph to support new scheduler
3 years ago
Christophe Favergeon
Richard Allen
SaiCharan
a1lu
Peter Torelli
Lingkai Dong
Silfurion
KoenVda88