OOICI Coverage Model

Initial coverage model implementation

Reference Information: https://confluence.oceanobservatories.org/display/CIDev/R2+Construction+Data+Model+Implementation

#Prerequisites

This assumes basic development environment setup (git, directory structure). Please follow the "New Developers Tutorial" for basic steps.

Install the following if not yet present:

OS Packages and package management: For Mac, use homebrew

/usr/bin/ruby -e "$(curl -fsSL https://raw.github.com/gist/323731)"

• git
• python 2.7
• hdf5
• netcdf

Install git, python, hdf5 and netcdf with Homebrew

brew install git python hdf5 netcdf

You'll also need the various dependencies of the pyon project:

brew install libevent libyaml zeromq couchdb rabbitmq pkg-config

You can even reinstall git using brew to clean up your /usr/local directory Be sure to read the pyon README for platform specific guidance to installing dependent libraries and packages. Linux: Note that many installs have much older versions installed by default. You will need to upgrade couchdb to at least 1.1.0.

Python packages and environment management:

Install pip

easy_install pip

Install virtualenv and virtualenvwrapper modules for your python 2.7 installation Note: This may require Mac's XCode (use XCode 3.3 free version

easy_install --upgrade virtualenv
easy_install --upgrade virtualenvwrapper

Setup a virtualenv to run coverage-model (use any name you like):

mkvirtualenv --python=python2.7 coverage_model
workon coverage_model
pip install numpy==1.6.2

Optional: To run any of the example functions that result in graphical output, you must also install matplotlib

pip install matplotlib

#Source

Obtain the coverage-model project by running:

git clone [email protected]:ooici/coverage-model.git
cd coverage-model

#Installation Ensure you are in a virtualenv prior to running the steps below

NOTE: The repository uses submodules for dependent OOI-CI projects. For more information on submodules, refer here

From the coverage-model directory, run the following commands:

git submodule init
git submodule update
python bootstrap.py
bin/buildout
bin/generate_interfaces

Once those steps complete, you should be able to import the coverage model from the python shell:

bin/pycc # or bin/ipython
from coverage_model import *

#Running unit tests (via nose)

From the coverage-model directory, run the following command:

bin/nosetests -v

This will run all UNIT and INT tests for the coverage-model repository. The -v flag denotes verbose output (the name of each test prints as it runs). For more nose options, refer to the nose documentation

#Using the coverage_model.test.examples functions

The coverage_model/test/examples.py module contains numerous functions that showcase the functionality of the Coverage Model - including generation of exemplar coverages, usage of ParameterTypes and more.
NOTE: The functions in this module are NOT guaranteed to work at all times!!

Start a pycc or ipython shell session from the root coverage-model directory:

cd your/code/path/coverage-model
bin/pycc # or bin/ipython

Next, simply import functions from the module and then try them out!

For example:

from coverage_model.test.examples import *
sample_cov = samplecov() # Generates a sample coverage with a few simple parameters
print sample_cov

ptypes_cov = ptypescov() # Generates a coverage with a parameter of each of the parameter types
print grid_cov

#Saving and Loading coverages

By default, the Coverage Model automatically persists data to disk using HDF5. The "top level" of a coverage is a directory. When loading a coverage, always refer to this top level directory.
Coverages can be loaded in two ways. Given the following directory structure:

.../my_data
|---ooi_coverages
    |---cov_one
        |--- ...
    |---cov_two
        |--- ...
    |---cov_three
        |--- ...

The coverage cov_one can be loaded by:

# Load method 1
cov = SimplexCoverage.load('.../my_data/ooi_coverages/cov_one')

# Load method 2
cov = SimplexCoverage('.../my_data/ooi_coverages', 'cov_one')

It is also possible to save coverages to single files using python pickle:
WARNING - DEPRECATED: pickle saving is not guaranteed to work and will be removed in future versions
NOTE: In order to save a coverage using this method, it MUST be an in-memory coverage (see below)

cov = SimplexCoverage.load('.../my_data/ooi_coverages/cov_one')

# Save to a pickle file
cov.pickle_save('.../my_data/ooi_coverages/pickled_cov_one.cov')

# Load from a pickle file
cov = SimplexCoverage.pickle_load('.../my_data/ooi_coverages/pickled_cov_one.cov')

#In-memory Coverage

Coverages can be created that do not automatically persist to HDF5 by setting the in_memory_storage=True when constructing the coverage.
NOTE: Using the Coverage Model in this manner can result in loss of information and should be used sparingly and/or only in certain situations

from coverage_model import *

# Assuming appropriate ParameterDictionary (pdict) and Domain (tdom) objects exist
cov = SimplexCoverage('test_data', 'mycov', 'my_coverage_name', parameter_dictionary=pdict, temporal_domain=tdom, in_memory_storage=True)