Nothing Special   »   [go: up one dir, main page]
Scribd Logo
Upload

Welcome to Scribd!

  • 3 views

    Model Asset Exchange (MAX) and The Data Asset Exchange (DAX) IBM Developer

    Uploaded by

    suvratml21

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Model Asset Exchange (MAX) and The Data Asset Exchange (DAX) IBM Developer

    Uploaded by

    suvratml21
    3 views7 pages

    Original Title

    Model Asset Exchange (MAX) and the Data Asset Exchange (DAX) IBM Developer

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    3 views7 pages

    Model Asset Exchange (MAX) and The Data Asset Exchange (DAX) IBM Developer

    Uploaded by

    suvratml21

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 7

    Getting started with the Model Asset Exchange and the Data

    Asset Exchange
    In this lab, you will explore the Model Asset Exchange (MAX) and the Data Asset Exchange (DAX), which are two open source Data
    Science resources on IBM Developer.

    Objective of Exercise 1:
    Find open data sets on IBM Developer.
    Explore the data sets.

    Objective of Exercise 2:
    Find ready-to-use deep learning models on the Model Asset Exchange.
    Explore the deep learning model trained to detect objects in an image.

    It will take you approximately 15 minutes to complete the lab. Only a web browser is required to complete the tasks.

    Exercise 1: Explore deep learning datasets


    The Data Asset Exchange is a curated collection of open datasets from IBM Research and third-parties that you can use to train models.

    1. Open https://developer.ibm.com/ in your web browser.

    2. From the main menu select Open Source at IBM > Data Asset eXchange. The DAX home page is displayed.

    The collection includes datasets from the Debater project (https://www.research.ibm.com/artificial-intelligence/project-debater/), datasets
    that can be used to train models to perform document layout analysis, natural language processing, time series analysis, and more.

    3. Open the NOAA Weather Data dataset (https://developer.ibm.com/exchanges/data/all/jfk-weather-data/), which contains data from
    a weather station at the John F. Kennedy Airport in New York spanning eight years.
    The dataset was published under the data science friendly CDLA-Sharing license (https://cdla.io/).
    The dataset contains time-series data and can be used to predict weather trends.
    This dataset was used to train the weather forecaster model on MAX (https://developer.ibm.com/exchanges/models/all/max-weather-
    forecaster/).

    4. Inspect the dataset's metadata.

    This dataset is stored as tabular data and formatted as a comma separated value (CSV) file, which is a very popular basic data exchange
    format.

    You can download the dataset using the Get this dataset link. Datasets are stored as compressed archives, which you can extract using
    any utility that supports the targz format. If you are not familiar with this file format, take a look at this short open source tutorial
    https://opensource.com/article/17/7/how-unzip-targz-file.

    5. Most datasets are complemented by Python notebooks that you can use to explore, pre-process, and analyze the data. The
    notebooks are hosted on Watson Studio, IBM's Data Science platform. Later in this course, you'll learn more about Watson Studio
    notebooks, how to sign up and how to run notebooks on them.

    For now, you can preview the dataset and the notebook (or notebooks) by clicking the Preview the data and notebooks as shown in the
    screenshot below.

    This will display dataset metadata, sample records and glossary as shown in the screenshot below.
    Now here, you click on the Notebook Preview option on top to view the notebook hosted with this dataset. Explore the steps followed in
    this notebook to clean data before performing data analysis on this dataset.

    This concludes Exercise 1 of this lab, which introduced the Data Asset Exchange. You may proceed to Exercise 2.

    6. [Optional] If you have already registered and are comfortable working with notebooks and Watson Studio, you can open the link
    using the Run Notebook in Watson Studio option and import it into a project. If you are not acquainted with IBM Watson Studio,
    you can skip this step. Detailed guidance on signing up and getting started with IBM Watson Studio will be provided in Week 7 of
    this course.

    Exercise 2 - Explore deep learning models


    The Model Asset Exchange is a curated repository of open source deep learning models for a variety of domains, such as text, image,
    audio, and video processing.

    For more details, please visit - https://github.com/CODAIT/max-central-repo webpage.

    The curated list includes deployable models, which you can run as a microservice locally or in the cloud on Docker or Kubernetes, and
    trainable models where you can use your own data to train the models. Some of the models are already built for you to test. Let's test one
    of the models.

    In this exercise, we explore the Object Detector model hosted on CodePen platform. This model recognizes the objects present in an
    image. The model consists of a deep convolutional net base model for image feature extraction, together with additional convolutional
    layers specialized for the task of object detection, trained on the COCO data set. The input to the model is an image, and the output are
    extracted objects from the image, appropriately labeled.

    CodePen is a social development environment. At its heart, it allows you to write code in the browser and see the results of it as you
    build. It is a useful and liberating online code editor for developers of any skill and is particularly empowering for people learning to
    code.

    1. Scroll down in the webpage and click CodePen as shown in the screenshot below.
    2. Select MAX TFJS models. Here we are using Image Segmenter, which divides an image into regions or categories that
    correspond to different objects or parts of objects. Every pixel in an image is allocated to one of a number of these categories.

    3. Click on Select Image and upload an image. You may choose images with a person, dog, cat, truck, car, and so on, which are
    labels the model has been trained on.
    Click here for all the labels the model is trained on

    4. Click the icon Extract prediction as shown below:


    You will now be able to see the output of the prediction on the basis of the image you upload.
    Here the background and the dog image are separated, showing two different parts of the image. You can also try the webcam option,
    which will show the real-time prediction by the toggle-on webcam option.

    This concludes Exercise 2 of this lab, which introduced the Model Asset Exchange.

    Optionally you can watch a demo of the Object detector model here.

    Author(s)
    Joseph Santarcangelo

    Other Contributor(s)

    Lavanya

    You might also like