The automatic annotation function malfunctions after integrating label-studio-ml-backend and segment_anything_model.

[chunxueer]

Holle . The automatic annotation function malfunctions after integrating label-studio-ml-backend and segment_anything_model.

1、Deployed label-studio-ml-backend and segment_anything_model

services:
  segment_anything_model:
    container_name: segment_anything_model
    image: heartexlabs/label-studio-ml-backend:sam-master
    init: true
    build:
      context: .
      shm_size: '4gb'
      args:
        TEST_ENV: ${TEST_ENV}
    deploy:
      resources:
        limits:
            memory: 8G
        reservations:
            memory: 4G
# Add this to pass through 1 GPU 
            devices:
              - driver: nvidia
                count: 1
                capabilities: [gpu]
    environment:
      # specify these parameters if you want to use basic auth for the model server
      - BASIC_AUTH_USER=
      - BASIC_AUTH_PASS=
      # Change this to your model name: MobileSAM or SAM
      - SAM_CHOICE=MobileSAM
      - LOG_LEVEL=DEBUG
      # Enable this to use the GPU
      - NVIDIA_VISIBLE_DEVICES=all
      # specify the number of workers and threads for the model server
      - WORKERS=1
      - THREADS=8
      # specify the model directory (likely you don't need to change this)
      - MODEL_DIR=/data/models

      # Specify the Label Studio URL and API key to access
      # uploaded, local storage and cloud storage files.
      # Do not use 'localhost' as it does not work within Docker containers.
      # Use prefix 'http://' or 'https://' for the URL always.
      # Determine the actual IP using 'ifconfig' (Linux/Mac) or 'ipconfig' (Windows).
      - LABEL_STUDIO_HOST=http://10.20.16.79:35001/
      - LABEL_STUDIO_ACCESS_TOKEN=0e3428a1c8403eae227fbb75df7101957f5e5c38

    ports:
      - 35002:9090
    volumes:
      - "./data/server:/data"

2、Configured in Settings -> Model, with the status shown as Connected

3、Configured in Settings -> Labeling Interface
(Only BrushLabels supported?)

<View>
  <Image name="image" value="$image" zoom="true"/>
  <BrushLabels name="tag" toName="image">
    <Label value="Airplane" background="rgba(255, 0, 0, 0.7)"/>
    <Label value="Car" background="rgba(0, 0, 255, 0.7)"/>
  </BrushLabels>
</View>

4、During annotation, no automatic segmentation occurs.(When clicking and brushing on the image, the label-studio-ml-backend does not print logs, nor does it receive any requests?)(When clicking on the task, the label-studio-ml-backend prints logs as expected.)

label-studio docker-compose.yml

services:
  sti-label-studio:
    stdin_open: true
    tty: true 
    image: heartexlabs/label-studio:latest
    restart: unless-stopped
    expose:
      - "8000"
    ports:
      - "35001:8000" 
    environment:
      - DJANGO_DB=default
      - POSTGRE_NAME=${POSTGRE_NAME}
      - POSTGRE_USER=${POSTGRE_USER}
      - POSTGRE_PASSWORD=${POSTGRE_PASSWORD}
      - POSTGRE_PORT=${POSTGRE_PORT}
      - POSTGRE_HOST=${POSTGRE_HOST}
      - LABEL_STUDIO_HOST=${LABEL_STUDIO_HOST:-}
      
      - AWS_ACCESS_KEY_ID=${AWS_ACCESS_KEY_ID}
      - AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY}
      - AWS_STORAGE_BUCKET_NAME=${AWS_STORAGE_BUCKET_NAME}
      - AWS_S3_ENDPOINT_URL=${AWS_S3_ENDPOINT_URL}
      - AWS_S3_REGION_NAME=${AWS_S3_REGION_NAME}
      - AWS_DEFAULT_ACL=${AWS_DEFAULT_ACL}
      - USE_S3_STORAGE=${USE_S3_STORAGE}

      - STORAGE_TYPE=${STORAGE_TYPE}
      - STORAGE_AWS_ACCESS_KEY_ID=${STORAGE_AWS_ACCESS_KEY_ID}
      - STORAGE_AWS_SECRET_ACCESS_KEY=${STORAGE_AWS_SECRET_ACCESS_KEY}
      - STORAGE_AWS_BUCKET_NAME=${STORAGE_AWS_BUCKET_NAME}
      - STORAGE_AWS_FOLDER=${STORAGE_AWS_FOLDER}
      - STORAGE_AWS_REGION_NAME=${STORAGE_AWS_REGION_NAME}
      - STORAGE_AWS_ENDPOINT_URL=${STORAGE_AWS_ENDPOINT_URL} 
      - STORAGE_AWS_OBJECT_PARAMETERS=${STORAGE_AWS_OBJECT_PARAMETERS}
      
      - JSON_LOG=1
    #      - LOG_LEVEL=DEBUG
    volumes:
      - /data/sti/label-studio/data:/label-studio/data:rw
    command: label-studio-uwsgi

Thanks a lot, looking forward to hearing from you.

[heidi-humansignal]

Hello,

I'm assuming you are following this documentation: https://github.com/HumanSignal/label-studio-ml-backend/tree/master/label_studio_ml/examples/segment_anything_model/

Thank you,
Abu

Comment by Abubakar Saad
Workflow Run