Quickstart guide (pip)¶
Welcome to SuperNNova! This is a quick start guide so you can start testing our framework. This guide assumes you have installed it with pip, if you want to use the GitHub cloning please refer to Quickstart guide (GitHub). Pip uses the master branch (not the frozen paper one and may be behind some updates).
pip install supernnova
Please beware that SuperNNova only runs properly in Unix systems.
Setup your environment. 3 options¶
For quick tests, a database that contains a limited number of light-curves is provided. It is located in
tests/raw. For more information on the available data, check Data walkthrough. An example of running as module can be found in
Build the database¶
In the parent folder, where
run.py is located you can launch python or ipython with the following:
import supernnova.conf as conf from supernnova.data import make_dataset # get config args args = conf.get_args() # create database args.data = True # conf: making new dataset args.dump_dir = "tests/dump" # conf: where the dataset will be saved args.raw_dir = "tests/raw" # conf: where raw photometry files are saved args.fits_dir = "tests/fits" # conf: where salt2fits are saved settings = conf.get_settings(args) # conf: set settings make_dataset.make_dataset(settings) # make dataset
Train an RNN¶
import supernnova.conf as conf from supernnova.training import train_rnn # get config args args = conf.get_args() args.train_rnn = True # conf: train rnn args.dump_dir = "tests/dump" # conf: where the dataset is saved args.nb_epoch = 2 # conf: training epochs settings = conf.get_settings(args) # conf: set settings train_rnn.train(settings) # train rnn
Validate an RNN¶
import supernnova.conf as conf from supernnova.validation import validate_rnn # get config args args = conf.get_args() args.validate_rnn = False # conf: validate rnn args.dump_dir = "tests/dump" # conf: where the dataset is saved settings = conf.get_settings(args) # conf: set settings validate_rnn.get_predictions(settings) # classify test set