tum-ei-eda/muriscv-nn
muRISCV-NN is a collection of efficient deep learning kernels for embedded platforms and microcontrollers.
C++Apache-2.0
Issues
- 1
Migrate to latest Toolchains
#59 opened by PhilippvK - 0
Support XLEN=64 in benchmarks & tests
#60 opened by PhilippvK - 0
- 0
Evaluate KERNELS_OPTIMIZED_FOR_SIZE kernels
#94 opened by PhilippvK - 11
- 0
Integration & Upstreaming of Vicuna/Capybara
#93 opened by PhilippvK - 0
- 0
Document workarounds to support VEXT on Vicuna
#31 opened by PhilippvK - 0
VWW model does not fit on Vicuna
#30 opened by PhilippvK - 4
tflm no commit specified
#17 opened by christian-herber-nxp - 2
- 2
RV64 support + baremetal binaries
#73 opened by jerryz123 - 1
muriscv_nn_avgpool_s16 channel loop
#18 opened by christian-herber-nxp - 0
Tracking Issue: LSTM Support
#23 opened by PhilippvK - 1
PEXT is getting a major refactoring
#28 opened by PhilippvK - 0
RVP: Replace staturating multiply-accumulate with non-saturating one to match CMSIS-NN
#46 opened by PhilippvK - 0
- 1
- 0
sync unit tests from cmsis-nn repo
#51 opened by PhilippvK - 0
- 0
Update README
#57 opened by PhilippvK - 0
- 2
How to generate model data?
#68 opened by fumao13579 - 0
Compare muRISCV-NN Performance with NMSIS
#69 opened by PhilippvK - 1
Spike: Latest RISC-V PK needs newer GCC release
#75 opened by PhilippvK - 11
Toolchain issues
#82 opened by chchenl - 1
Port batch_norm kernels
#86 opened by PhilippvK - 0
Refactor muriscv_nn_vector_sum_s8
#87 opened by PhilippvK - 0
- 0
RVV Kernel in muriscv_nn_vec_mat_mult_t_s8 does not support address_offset != 1
#81 opened by PhilippvK - 0
- 0
Migrate to CMSIS-NNs new test infrastructure
#84 opened by PhilippvK - 1
Implement new transpose kernels
#85 opened by PhilippvK - 0
- 0
- 0
- 1
RVV Intrinsic Error with new RISC-V GCC
#77 opened by PhilippvK - 0
Fix CI deprecations
#63 opened by PhilippvK - 0
Align used Ubuntu Version in CI Workflows
#62 opened by PhilippvK - 0
- 0
Fix README batches
#36 opened by PhilippvK - 6
Fix broken compatibility with upstream TFLM
#22 opened by PhilippvK - 5
Questions on the difference between the ARM and muRISCV implementations in nn_mat_mult_kernel_s8_s16
#64 opened by tomhepworth - 2
- 0
add muriscvnn paper bitex entry to repo
#53 opened by PhilippvK - 1
Use updated Intrinsics API for V-EXT
#29 opened by PhilippvK - 3
- 0
muriscv_elementwise_mul_s16_s8.c vs muriscv_nn_elementwise_mul_s16_s8.c
#33 opened by christian-herber-nxp - 1
Spike TFLM integration tests broke
#25 opened by PhilippvK - 1
Refactor Unit Tests (Fix OVPSim)
#24 opened by PhilippvK