Skip to content

Latest commit

 

History

History
49 lines (34 loc) · 1.9 KB

quickstart.md

File metadata and controls

49 lines (34 loc) · 1.9 KB

Quick Start

  1. Install loom following the Install Guide and activate the virtualenv you created during the install process, for example using:

    workon loom
    
  2. (optional) Set up a remote ipython notebook server

    ssh -A -L 8888:localhost:8888 <user>@<hostname>
    workon loom
    pip install ipython[all]
    pip install --upgrade pyzmq
    ipython notebook --no-browser --ip=0.0.0.0 --matplotlib=inline
    

    You should now be able to access the ipython notebook server at http://localhost:8888

  3. There are three basic steps in the loom workflow: preparing and ingesting data, running inference, and querying the results.

    To prepare a dataset for ingestion, loom needs two files: the data in a csv file, and a schema indicating which feature models to use for each column in the csv.

    The taxi example contains both of these, ready to go: example.csv, schema.json.

  4. Ingest the data. Loom will read in the csv and the schema, and translate them into its own on-disk representation.

    You must supply a name, which is how loom will refer to this analysis in subsequent steps; we will use the name "quickstart".

    cd $LOOM/examples/taxi
    python -m loom.tasks ingest quickstart schema.json example.csv
    
  5. Run inference. This step reads the ingested data from the previous step, and produces indexes that can be queried.

    python -m loom.tasks infer quickstart
    
  6. Interactively query loom using the client library. See here for more information about the supported query operations.

    python
    import loom.tasks
    with loom.tasks.query("quickstart") as server:
        related = server.relate(["feature1", "feature2", "feature3"])
        print related