[Detector Support]: RKNN_QUERY_INPUT_DYNAMIC_RANGE error

[3AXAPOB]

Describe the problem you are having

Object detection does not work, this is in the logs:
2024-06-22 23:43:08.196220696 I RKNN: [23:43:08.196] RKNN Model Information, version: 6, toolkit version: 1.6.0+81f21f4d(compiler version: 1.6.0 (585b3edcf@2023-12-11T08:03:14)), target: RKNPU v2, target platform: rk3588, framework name: ONNX, framework layout: NCHW, model inference type: static_shape
2024-06-22 23:43:08.212846208 W RKNN: [23:43:08.212] query RKNN_QUERY_INPUT_DYNAMIC_RANGE error, rknn model is static shape type, please export rknn with dynamic_shapes
2024-06-22 23:43:08.213031704 W Query dynamic range failed. Ret code: RKNN_ERR_MODEL_INVALID. (If it is a static shape RKNN model, please ignore the above warning message.)

Version

0.13.2-69D9A261

Frigate config file

mqtt:
  enabled: false
detectors:
  rknn:
    type: rknn
    core_mask: 0b111
model:
  path: default-yolov8n
  width: 320
  height: 320
  input_tensor: nhwc
  input_pixel_format: bgr

rtmp:
  enabled: false

cameras:
  Fasad3: # <--- this will be changed to your actual camera later
    enabled: true
    birdseye:
      order: 2
    ffmpeg:
      input_args: preset-rtsp-restream
      hwaccel_args: preset-rk-h264
      inputs:
      - path: rtsp:https://admin:[email protected]:554?channel=0&subtype=0
        roles:
        - detect
    detect:
      enabled: True
      width: 1280
      height: 720
      fps: 5
    objects:
      track:
      - person

docker-compose file or Docker CLI command

services:
    frigate:
      container_name: frigate
      image: marca711/frigate:latest-rk
      privileged: true # this may not be necessary for all setups
      restart: unless-stopped
      group_add:
        - "106" # render
        - "44"  # video
        - "46"  # plugdev
      shm_size: "512mb" # update for your cameras based on calculation above
      devices:
        - /dev/dri/renderD128:/dev/dri/renderD128
        - /dev/dri/card0:/dev/dri/card0
        - /dev/rga
        - /dev/video-dec0:/dev/video-dec0
        - /dev/video-enc0:/dev/video-enc0
      volumes:
        - /dev/bus/usb:/dev/bus/usb
        - /etc/localtime:/etc/localtime:ro
        - /root/frigate/config.yml:/config/config.yml:rw
        - /root/frigate/storage:/media/frigate:rw
        - type: tmpfs # Optional: 1GB of memory, reduces SSD/SD Card wear
          target: /tmp/cache
          tmpfs:
            size: 2000000000
      ports:
        - "5000:5000"
        - "1935:1935"
        - "8554:8554" # RTSP feeds
        - "8555:8555/tcp" # WebRTC over tcp
        - "8555:8555/udp" # WebRTC over udp
      device_cgroup_rules:
        - "c 189:* rmw"

Relevant log output

2024-06-22 23:43:06.112901121  [INFO] Preparing Frigate...
2024-06-22 23:43:06.182514309  [INFO] Starting Frigate...
2024-06-22 23:43:07.910467980  [2024-06-22 23:43:07] frigate.app                    INFO    : Starting Frigate (0.13.2-69d9a261)
2024-06-22 23:43:07.929808557  [2024-06-22 23:43:07] peewee_migrate.logs            INFO    : Starting migrations
2024-06-22 23:43:07.934475997  [2024-06-22 23:43:07] peewee_migrate.logs            INFO    : There is nothing to migrate
2024-06-22 23:43:07.941167267  [2024-06-22 23:43:07] frigate.app                    INFO    : Recording process started: 453
2024-06-22 23:43:07.943840334  [2024-06-22 23:43:07] frigate.app                    INFO    : go2rtc process pid: 89
2024-06-22 23:43:07.971812015  [2024-06-22 23:43:07] frigate.app                    INFO    : Output process started: 463
2024-06-22 23:43:07.972225589  [2024-06-22 23:43:07] detector.rknn                  INFO    : Starting detection process: 462
2024-06-22 23:43:07.984773106  [2024-06-22 23:43:07] frigate.app                    INFO    : Camera processor started for Fasad3: 468
2024-06-22 23:43:07.991434044  [2024-06-22 23:43:07] frigate.app                    INFO    : Capture process started for Fasad3: 470
2024-06-22 23:43:08.195890245  I RKNN: [23:43:08.195] RKNN Runtime Information, librknnrt version: 1.6.0 (9a7b5d24c@2023-12-13T17:31:11)
2024-06-22 23:43:08.195912994  I RKNN: [23:43:08.195] RKNN Driver Information, version: 0.9.6
2024-06-22 23:43:08.196220696  I RKNN: [23:43:08.196] RKNN Model Information, version: 6, toolkit version: 1.6.0+81f21f4d(compiler version: 1.6.0 (585b3edcf@2023-12-11T08:03:14)), target: RKNPU v2, target platform: rk3588, framework name: ONNX, framework layout: NCHW, model inference type: static_shape
2024-06-22 23:43:08.212846208  W RKNN: [23:43:08.212] query RKNN_QUERY_INPUT_DYNAMIC_RANGE error, rknn model is static shape type, please export rknn with dynamic_shapes
2024-06-22 23:43:08.213031704  W Query dynamic range failed. Ret code: RKNN_ERR_MODEL_INVALID. (If it is a static shape RKNN model, please ignore the above warning message.)

Operating system

Debian

Install method

Docker Compose

Object Detector

RKNN

Any other information that may be helpful

No response

[NickM-27]

CC @MarcA711

[MarcA711]

Is something not working? As the error message itself says, it can be ignored.

[3AXAPOB]

Is something not working? As the error message itself says, it can be ignored

Yes, nothing detects.

[3AXAPOB]

The detector started working after reinstalling the container.
Thank you for your attention to my question!