WISE-PaaS Marketplace

Discovering and Shopping for Your Own IoT Deployments

WISE-PaaS/AFS
  • Service Provider:Advantech
  • Category : IoT Cloud Services, Industrial Cloud Solutions
  • Last Update: 9/25/2019 12:00:00 AM
AFS is designed to serve as multi-model training and multi-model deployment services.
{{vm.prdAction.Point}} up
Perpetual
Monthly Subscription
{{vm.prdAction.Point}} up
Monthly Subscription
{{vm.prdAction.Point}} up
Monthly Subscription
(Trial Available)
{{vm.prdAction.Point}} up
Monthly Subscription / Redeem
Limited
Advantech professional can help find the best solution for your business.
  • Service Details
    Introduction

    WISE-PaaS/AFS is designed to serve as an AI framework service for implementing data analytics with model training and model deployment services on the WISE-PaaS Industrial IoT cloud platform. Our WISE-PaaS/AFS provides a unified platform for AI model training, deployment, re-training, and re-deployment.


    AFS_Info.jpg


    Feature Highlights

    1. 1. Workspace:
      1. Online Code IDE: We provide the Jupyter Notebook to help our partners quickly develop many kinds of analytic models based on Python3. Machine learning and deep learning frameworks are also supported, such as Scikit-Learn, Pytorch, Keras and TensorFlow.
      2. Upload Code: If you want to program offline, you also can upload your code package and docker image as runtime.
    2. 2. Task Management:
      1. Training Tasks: Helps you easily set up a training task schedule to active job training automatically.
      2. Model Deployment Tasks: Helps you easily set up model deployment schedules and trigger rules, such as best model and model performance.
    3. 3. Catalog:
      1. With many prebuild analytic modules, all you have to do is subscribe and then start job training immediately.
    4. 4. Model Board:
      1. We provide model boards to visualize learning, evaluate models, and Web UI.
    5. 5. Inference Engine:
      1. Edge Inference: Helps you quickly deploy model remotely to edge devices without on-site installation effort.
      2. Cloud Inference: Helps you easily push inference engine app to your EnSaaS Org/Space and provides RestfulAPI automatically.
    6. 6. AFS SDK:
      1. Model SDK: Helps you easily upload your training models to AFS Model repository.
      2. Data Source SDK: Helps you easily insert, query, and update your training data in WISE-PaaS PostgeSQL, MongoDB, InfluxDB, Blob Service, and WISE-PaaS/APM.

    AFS_Feature01.png

    Pricing Details AFS_Feature03.png
  • Pricing Details
    {{category.PlanCategoryName}}
Introduction

WISE-PaaS/AFS is designed to serve as an AI framework service for implementing data analytics with model training and model deployment services on the WISE-PaaS Industrial IoT cloud platform. Our WISE-PaaS/AFS provides a unified platform for AI model training, deployment, re-training, and re-deployment.


AFS_Info.jpg


Feature Highlights

  1. 1. Workspace:
    1. Online Code IDE: We provide the Jupyter Notebook to help our partners quickly develop many kinds of analytic models based on Python3. Machine learning and deep learning frameworks are also supported, such as Scikit-Learn, Pytorch, Keras and TensorFlow.
    2. Upload Code: If you want to program offline, you also can upload your code package and docker image as runtime.
  2. 2. Task Management:
    1. Training Tasks: Helps you easily set up a training task schedule to active job training automatically.
    2. Model Deployment Tasks: Helps you easily set up model deployment schedules and trigger rules, such as best model and model performance.
  3. 3. Catalog:
    1. With many prebuild analytic modules, all you have to do is subscribe and then start job training immediately.
  4. 4. Model Board:
    1. We provide model boards to visualize learning, evaluate models, and Web UI.
  5. 5. Inference Engine:
    1. Edge Inference: Helps you quickly deploy model remotely to edge devices without on-site installation effort.
    2. Cloud Inference: Helps you easily push inference engine app to your EnSaaS Org/Space and provides RestfulAPI automatically.
  6. 6. AFS SDK:
    1. Model SDK: Helps you easily upload your training models to AFS Model repository.
    2. Data Source SDK: Helps you easily insert, query, and update your training data in WISE-PaaS PostgeSQL, MongoDB, InfluxDB, Blob Service, and WISE-PaaS/APM.

AFS_Feature01.png

Pricing Details AFS_Feature03.png

WISE-PaaS/AFS System Architecture:


AFS_Architecture02.png


Use Cases:


Steel Factory

Equipment anomaly detection for steel factories can be achieved by incorporating APM, AFS, and Dashboard. They implement feature extraction processes and upload features using SCADA SDK to the APM portal where asset profile and management logistics have been configured. After collecting and labeling enough datasets, AFS is used to choose or develop a training package and trigger the training task with the APM firehose setting. After training, model performance can be monitored and data can be relabeled directly on the dashboard.

When the model is ready, users can rely on AFS inference engine functionality to deploy the model and the corresponding inference Docker image to edge devices.

The results from inference execution can also be uploaded back to the APM portal where the event rule has been defined using APM Rule Engine to monitor the prediction results. When the event has been triggered, WISE-PaaS Notification Service informs the equipment operation and maintenance team to act accordingly.

{{category.PlanCategoryName}} See More

Related Services