CMSIS-Executorch/cmsis-executorch.csolution.yml at main · Arm-Examples/CMSIS-Executorch · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
# Copyright 2026 Arm Limited and/or its affiliates.
# SPDX-License-Identifier: Apache-2.0
#
# A template for ExecuTorch applications, initially targeting an Ethos-U85 on
# the Corstone-320 (SSE-320) FVP.
# Three steps build it:
#
#   cbuild setup cmsis-executorch.csolution.yml --active SSE-320-U85
#   python create_ai_layer.py cmsis-executorch.cbuild-mlops.yml
#   cbuild cmsis-executorch.csolution.yml --active SSE-320-U85
#
# See CMSIS-Toolbox MLOps information:
# https://open-cmsis-pack.github.io/cmsis-toolbox/build-overview/#mlops-information
# `cbuild setup` exports it for create_ai_layer.py to generate the AI layer.
solution:
  created-for: CMSIS-Toolbox@2.15.0
  cdefault:
  compiler: AC6

  packs:
    # The PyTorch::ExecuTorch pack version must match the ExecuTorch Python
    # exporter in requirements.txt because its runtime consumes the generated .pte.
    - pack: PyTorch::ExecuTorch@1.5.1
    - pack: ARM::Cortex_DFP

  target-types:
    # Corstone-320 with an Ethos-U85 NPU, simulated on the Arm FVP.
    - type: SSE-320-U85
      board: SSE-320
      device: SSE-320-FVP
      variables:
        - Board-Layer: $SolutionDir()$/board/Corstone-320/Board-U85.clayer.yml
        - AI-Layer: $SolutionDir()$/ai_layer/ai_layer.clayer.yml
      target-set:
        - set:
          images:
            - project-context: cmsis-executorch.Debug
          debugger:
            name: Arm-FVP
            # On macOS use .vscode/fvp.sh and copy launch.json.mac to launch.json.
            # See: https://github.com/Arm-Examples/FVPs-on-Mac/tree/main/skills/fvp-debug-setup
            model: FVP_Corstone_SSE-320
            config-file: board/Corstone-320/fvp_config.txt
            args: "--simlimit 60"

  build-types:
    - type: Debug
      debug: on
      optimize: debug

  projects:
    - project: ./cmsis-executorch.cproject.yml

  # Target and model settings consumed by create_ai_layer.py.
  mlops:
    description: TinyCNN int8 image classifier for Ethos-U85
    npu:
      type: Ethos-U85                # explicit: the SSE-320 device pack declares no NPU
    vela:
      system: Ethos_U85_SYS_DRAM_Mid # system-config from the Vela config
      memory: Shared_Sram            # memory-mode from the Vela config
    model:
      clayer: $AI-Layer$
      name: TinyCNN
      # Model parameters for create_ai_layer.py: the NHWC input shape and the
      # number of samples the quantizer is calibrated with.
      input-shape: 1x16x16x3
      calibration-samples: 2