Versions Compared

Old Version 151

changes.mady.by.user Smruti Padhy

Saved on

compared with

New Version Current

changes.mady.by.user Smruti Padhy

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Create applications using BD services (50 mins)

    • Conversion Example (15 mins20mins):  To convert a collection of images/ps/odp/audio/video files to png/pdf/ppt/mp3.  This will demonstrates that if you have a directory with files in old file formats, just use BD to get it all converted without requiring to install any software. (Emphasizes on conversion)
      • Make sure imagemagick and ffmpeg converters are running before the demo
      • Obtain the BD token/key - ask participant to refer to previous bdfiddle step or use the python library
      • Ask the participant to check for the available output formats for specific input formats
      • Ask the participant to use python library to use BD service
      • TODO:
        •  Inna Zharnitsky Provide a Python script for this and let participants use python library to use the  BD service
        •  Smruti PadhyProvide a Step-by-step instructions with screenshot to do this
        •  (Optional) Provide R script for this problem.

    • Extraction/Indexing/Retrieval Example (20 mins): 

      Given a collection of images with text embedded in it, and audio files, search images/audio files based on its content. (

      Emphasizes on extraction on unstructured data, indexing and content-based retrieval)

      • Make sure OCR, face, and speech2text extractors are running before starting the demo
      • One can upload images from local directory to obtain images or use external web service.  
      • Let the participant use the python library of BD to obtain key/token and submit extraction request to BD-API gateway
      • Once technical metadata is obtained from BD, write the tags and technical metadata to a local file /python dictionary.
      • Search the file based on the tags/technical metadata by linear search on the index file

      • TODO
        •  Make sure the metadata to be posted as json-ld for the extractors 
        •  Create an example dataset with images and audios to which we can make interesting query
        •  (Optional) Provide a code snippet of using externel service to obtain images. e.g. Flicker API.
            •  This will only be provided as an example and will not be used for the rest of the code.
        •  Provide the link to the current BD REST API and create a document/wiki page showing step-by-step screenshots of obtaining a key/token using python library.
        •  Marcus SlavenasWrite a Python script that will serve as a stub for the BD client
          • The participant will fill in the code to use python library to call BD REST API and submit their requests. 
          •  The python script  should write the tags and technical metadata to a local file. (Probably can use python library's index method that writes it as feature vectors.)
        •   Marcus SlavenasWrite a Python script to make interesting search/query to the index file. Again probably use the python library's find method or just read the local file.     
        •  (Optional) provide R script for this problem

    •  Conversion & Extraction Example (15 10 min):  Given an image file, convert it to a different format and obtains the face detection and OCR. 
      • TODO
        •  Write a Python script that does the conversion and then sends the converted file to the BD service.
        •  Create a step-by-step instructions document with screenshots.
        •  (Optional) Provide a R script for this problem

    • (Optional) Combination of Conversion & Extraction Example: Obtaining Ameriflux data and converting into *.clim format (similar to csv format but tab separated) for SNIPET model.  Calculate average air temperature and its standard deviation. (This will emphasize both conversion and analysis)

...

  • Part 1: Teach to write an extractor (35 mins)
    • Start with the bd-template extractor, which is the word count extractor.
    • Ask participant to modify the extractor, which would use 'grep' to find a specific pattern within the file.
    • Ask to change the name of the extractor from ncsa.wordcount to ncsa.grep.
    • Include yes/no in the metadata if the pattern is found or not found.
    • Briefly describe Json-ld support. Provide intuition behind the idea json-ld with a simple example. No need to go into details of RDF. 
    • TODO
      •  Smruti Padhy, Marcus Slavenas, Jong Lee  Provide Step-by-step instructions/screenshots of updating the extractor and the output as seen at the Clowder GUI.  Also provide link to json-ld for further readings. Provide minimum software requirements for the development such as Clowder, Rabbimq, MongoDB, pyclowder, python libraries, etc.
      •  Inna Zharnitsky Write an extractor that does grep along with the wordcount for demonstration purpose and include json-ld 
      •  Sandeep Puthanveetil Satheesan Write an extractor that accepts csv file with say 3 columns (probably with values from weather or bacterial growth model (see Problem 2.2 below)) , calculate the average of a specific column
        •  Provide step-by-step screenshots for writing such an extractor.

  • Part 2: Teach to write a converter (35 mins)
    • Start with the bd-template for converter- imagemagick
    • Ask the participant to modify the converter input/output formats in the comment section. And see the result using the polyglot web UI for post and get
    • Another example - FFmpeg converter for audio and video
    • TODO

  • Part 3: Teach to upload a converter or an extractor to the locally installed Tools Catalog. (20mins)
    •  Inna Zharnitsky Step-by-step procedure to upload an extractor or a converter, an input file and an output file without a docker file.

  • Part 4 (Optional - For advanced user): Dockerize the tool

...