Benchmarks 2024 02 20 TFLM GCC - tum-ei-eda/muriscv-nn GitHub Wiki
Setup
Simulator
- Spike (
riscv-isa-sim) (ISS, CPI=1)
Toolchains
- RISC-V GCC:
- Scalar: TODO: version & url
- Vector: TODO: version & url
- Packed: Self compiled using patches found in https://github.com/riscv-collab/riscv-gcc/pull/258 and https://github.com/riscvarchive/riscv-binutils-gdb/pull/257
Models
-
MLPerfTiny Benchmark
-
TODO: others!
Package Versions
-
MLonMCU : main
-
TFLM : main
-
Spike : 0bc176b3fca43560b9e8586cdbc41cfde073e17a
-
Spike PK : 7e9b671c0415dfd7b562ac934feb9380075d4aa2
Miscellaneous
- Used
-Osflag for compilation. - Benchmarks generated using MLonMCU deployment tool with minimal efforts.
- Memory metrics are reported in Bytes
Results (Framework: tflm, Backend: tflmi, Toolchain: gcc)
Audio Wake Words (aww)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Kernels | Mode | Arch | Auto-Vectorization |
|---|---|---|---|---|---|---|---|
174698646 ( 0.1x ) |
132407 ( 0.879 ) |
36204 ( 1.0 ) |
0 | TFLM | Reference | RV32GC | - |
174698646 ( 0.1x ) |
132413 ( 0.879 ) |
36204 ( 1.0 ) |
128 | TFLM | Reference | RV32GCV | Loop+SLP |
174698646 ( 0.1x ) |
132413 ( 0.879 ) |
36204 ( 1.0 ) |
1024 | TFLM | Reference | RV32GCV | Loop+SLP |
157549999 ( 0.1x ) |
144774 ( 0.961 ) |
36148 ( 0.998 ) |
0 | TFLM | Reference | RV32GCP | - |
16643414 ( Base ) |
150578 ( Base ) |
36212 ( Base ) |
0 | muRISCV-NN | Scalar | RV32GC | - |
16643414 ( 1.0x ) |
150584 ( 1.0 ) |
36212 ( 1.0 ) |
128 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
16643414 ( 1.0x ) |
150584 ( 1.0 ) |
36212 ( 1.0 ) |
1024 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
6995715 ( 2.4x ) |
151102 ( 1.003 ) |
36212 ( 1.0 ) |
128 | muRISCV-NN | Vector | RV32GCV | - |
2565027 ( 6.5x ) |
151102 ( 1.003 ) |
36212 ( 1.0 ) |
1024 | muRISCV-NN | Vector | RV32GCV | - |
13496950 ( 1.2x ) |
162164 ( 1.077 ) |
36156 ( 0.998 ) |
0 | muRISCV-NN | Scalar | RV32GCP | - |
15933233 ( 1.0x ) |
164564 ( 1.093 ) |
36156 ( 0.998 ) |
0 | muRISCV-NN | Packed | RV32GCP | - |
Image Classification (resnet)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Kernels | Mode | Arch | Auto-Vectorization |
|---|---|---|---|---|---|---|---|
745801113 ( 0.1x ) |
172997 ( 0.935 ) |
68968 ( 1.0 ) |
0 | TFLM | Reference | RV32GC | - |
745801113 ( 0.1x ) |
173013 ( 0.935 ) |
68968 ( 1.0 ) |
128 | TFLM | Reference | RV32GCV | Loop+SLP |
745801113 ( 0.1x ) |
173013 ( 0.935 ) |
68968 ( 1.0 ) |
1024 | TFLM | Reference | RV32GCV | Loop+SLP |
697912970 ( 0.1x ) |
185266 ( 1.001 ) |
68912 ( 0.999 ) |
0 | TFLM | Reference | RV32GCP | - |
80992847 ( Base ) |
185078 ( Base ) |
68960 ( Base ) |
0 | muRISCV-NN | Scalar | RV32GC | - |
80992847 ( 1.0x ) |
185110 ( 1.0 ) |
68960 ( 1.0 ) |
128 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
80992847 ( 1.0x ) |
185110 ( 1.0 ) |
68960 ( 1.0 ) |
1024 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
29958029 ( 2.7x ) |
186434 ( 1.007 ) |
68960 ( 1.0 ) |
128 | muRISCV-NN | Vector | RV32GCV | - |
8345189 ( 9.7x ) |
186434 ( 1.007 ) |
68960 ( 1.0 ) |
1024 | muRISCV-NN | Vector | RV32GCV | - |
62969349 ( 1.3x ) |
196768 ( 1.063 ) |
68904 ( 0.999 ) |
0 | muRISCV-NN | Scalar | RV32GCP | - |
68428919 ( 1.2x ) |
199762 ( 1.079 ) |
68904 ( 0.999 ) |
0 | muRISCV-NN | Packed | RV32GCP | - |
Anomaly Detection (toycar)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Kernels | Mode | Arch | Auto-Vectorization |
|---|---|---|---|---|---|---|---|
3094956 ( 0.6x ) |
333908 ( 0.989 ) |
19432 ( 1.0 ) |
0 | TFLM | Reference | RV32GC | - |
3094956 ( 0.6x ) |
333914 ( 0.989 ) |
19432 ( 1.0 ) |
128 | TFLM | Reference | RV32GCV | Loop+SLP |
3094956 ( 0.6x ) |
333914 ( 0.989 ) |
19432 ( 1.0 ) |
1024 | TFLM | Reference | RV32GCV | Loop+SLP |
3097895 ( 0.6x ) |
346264 ( 1.026 ) |
19380 ( 0.997 ) |
0 | TFLM | Reference | RV32GCP | - |
1969300 ( Base ) |
337550 ( Base ) |
19432 ( Base ) |
0 | muRISCV-NN | Scalar | RV32GC | - |
1969300 ( 1.0x ) |
337556 ( 1.0 ) |
19432 ( 1.0 ) |
128 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
1969300 ( 1.0x ) |
337556 ( 1.0 ) |
19432 ( 1.0 ) |
1024 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
608151 ( 3.2x ) |
337804 ( 1.001 ) |
19432 ( 1.0 ) |
128 | muRISCV-NN | Vector | RV32GCV | - |
428279 ( 4.6x ) |
337804 ( 1.001 ) |
19432 ( 1.0 ) |
1024 | muRISCV-NN | Vector | RV32GCV | - |
1873278 ( 1.1x ) |
349712 ( 1.036 ) |
19380 ( 0.997 ) |
0 | muRISCV-NN | Scalar | RV32GCP | - |
942627 ( 2.1x ) |
351430 ( 1.041 ) |
19380 ( 0.997 ) |
0 | muRISCV-NN | Packed | RV32GCP | - |
Visual Wake Words (vww)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Kernels | Mode | Arch | Auto-Vectorization |
|---|---|---|---|---|---|---|---|
495266208 ( 0.1x ) |
406111 ( 0.957 ) |
134520 ( 1.0 ) |
0 | TFLM | Reference | RV32GC | - |
495266207 ( 0.1x ) |
406117 ( 0.957 ) |
134520 ( 1.0 ) |
128 | TFLM | Reference | RV32GCV | Loop+SLP |
495266207 ( 0.1x ) |
406117 ( 0.957 ) |
134520 ( 1.0 ) |
1024 | TFLM | Reference | RV32GCV | Loop+SLP |
445892345 ( 0.1x ) |
418478 ( 0.986 ) |
134464 ( 1.0 ) |
0 | TFLM | Reference | RV32GCP | - |
49673130 ( Base ) |
424282 ( Base ) |
134528 ( Base ) |
0 | muRISCV-NN | Scalar | RV32GC | - |
49673130 ( 1.0x ) |
424288 ( 1.0 ) |
134528 ( 1.0 ) |
128 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
49673130 ( 1.0x ) |
424288 ( 1.0 ) |
134528 ( 1.0 ) |
1024 | muRISCV-NN | Scalar | RV32GCV | Loop+SLP |
21934482 ( 2.3x ) |
424806 ( 1.001 ) |
134528 ( 1.0 ) |
128 | muRISCV-NN | Vector | RV32GCV | - |
10505033 ( 4.7x ) |
424806 ( 1.001 ) |
134528 ( 1.0 ) |
1024 | muRISCV-NN | Vector | RV32GCV | - |
40746534 ( 1.2x ) |
435868 ( 1.027 ) |
134472 ( 1.0 ) |
0 | muRISCV-NN | Scalar | RV32GCP | - |
49175307 ( 1.0x ) |
438268 ( 1.033 ) |
134472 ( 1.0 ) |
0 | muRISCV-NN | Packed | RV32GCP | - |
Original data
Click here to download the raw files for this benchmark.