An I/O HDF5-based utility to assist in the distribution, execution and integration of data coming from different sources (e.g. results of mathematical optimization models).
Project description
dirB
A generic tool working over HDF5 files to share experiments data, both input and output, in a standardized way.
JSON dictionaries are the default objects used within the HDF5 files in order to collect input parameters and resulting data.
Features
The main features available in dirB are:
- Create new cases with input parameters and metadata (
dirB.base.creaNuevoCaso(inputParamsDict, metadataDict)
) - Load already existing cases from HDF5 files (
dirB.base.cargaCaso(hdf5FileName)
) - Merge several files containing solutions (
dirB.merge.mergeListaDeSoluciones(baseCase, listOfFilesWithSolutions, outputFile)
)
Features to be implemented:
- Retrieval of solutions fulfilling a certain criteria, in order to process all them together
Examples
To create a new case from scratch:
from dirB.base import zsan_DirB
from datetime import date, datetime
import time
import shutil
diccionarioCasoBase = { 'x': 12,
'y': 7}
diccionarioAtributosCasoBase = {
'Autor': 'Steve Vai',
'Dia': str(date.today()),
'Escenario': 'Acido',
'Numero Caso': 23
}
casoBase = zsan_DirB()
casoBase.creaNuevoCaso(diccionarioCasoBase, diccionarioAtributosCasoBase, nombreDeFichero = "casoBase.hdf5")
For new partners looking to run this case within their own environments, the first step is to copy the file. Then, run the model with the considered solvers and save them into regular Python dictionaries, which should be added to the new file:
shutil.copy("casoBase.hdf5", "casoConSolucion_01.hdf5")
time.sleep(2)
# Assume these are the 2 resulting dictionaries
diccionarioSolucionCaso_1 = {'ex1': 'xxx',
'ex2': 'yyy',
'ex3': 'zzz'}
diccionarioAtributosSolucionCaso_1 = {
'Autor': 'Michel Gondry',
'Dia': str(datetime.now()),
'Escenario': 'Amargo',
}
solucion_1 = zsan_DirB()
solucion_1.cargaCaso("casoConSolucion_01.hdf5")
solucion_1.guardaNuevaSolucion(diccionarioSolucionCaso_1, diccionarioAtributosSolucionCaso_1)
For every new experiment, the same process should be followed.
Finally, when all the runs have been finished, they should be added up into the same file. As an example, consider we've got a total of 3 runs and that we want to merge them all into our base case file. An additional parameter should be provided to specify the resulting file name, as the base case one will never be overwritten by this method.
from dirB.merge import mergeListaDeSoluciones
casoBase = 'casoBase.hdf5'
listaDeCasos = ['casoConSolucion_01.hdf5', 'casoConSolucion_02.hdf5', 'casoConSolucion_03.hdf5']
id_salida = "salida.hdf5"
mergedList = mergeListaDeSoluciones(casoBase, listaDeCasos, id_salida)
Complete example notebooks have been added to this repository.
Documentation
Here is the official documentation. For any modification you'd like to do it, it can be generated locally. In order to do this, we recommend to create a brand new environment for sphinx, in which you will install the proper dependencies by means of pip install docs/{requirements_file}
:
- docs/requirements.in: contains the base dependencies
- docs/requirements.txt: contains the base dependencies plus the whole bunch of subdependencies recursively gotten for complete reproducibility
Then, you should run the make
command for the HTML creation:
cd docs
make html
With the following command we'll get a clean documentation env to regenerate the whole documentation in case we need to:
make clean html
make html
Acknowledgements
© Copyright 2023, Germán Navarro $^\dagger$, Santiago Fernández Prieto $^{\ddagger,1}$, Ricardo Enríquez Miranda $^\ddagger$, Javier Hernanz Zájara $^\ddagger$
$^\dagger$ Barcelona Supercomputing Center
$^\ddagger$ Repsol Quantum Advisory Team
$^1$ Repsol-BSC Research Center
Developed within the framework of the project CUCO. Financed by the CDTI and with the support of the Spanish Ministry of Science and Innovation under the Recovery, Transformation and Resilience Plan (RTRP).
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.