Skip to content

Latest commit

 

History

History
69 lines (46 loc) · 1.75 KB

README.md

File metadata and controls

69 lines (46 loc) · 1.75 KB

Light curve classifier (LCC)

Light curve classifier (LCC) is an approach to classify lightcurves using convolutional neural networks.

Install LCC

Clone Repository with Git

Clone the LCC repository:

cd </installation/path>
git clone https://github.com/TjarkMiener/LightCurveClassifier.git

Install Package with Anaconda

Next, download and install Anaconda, or, for a minimal installation, Miniconda. Create a new conda environment that includes all the dependencies for LCC:

conda env create -f </installation/path>/LightCurveClassifier/environment-<MODE>.yml

where <MODE> is either 'cpu' or 'gpu', denoting the TensorFlow version to be installed. If installing the GPU version of TensorFlow, verify that your system fulfills all the requirements here.

Finally, install LCC into the new conda environment with pip:

source activate lcc
cd </installation/path>/LightCurveClassifier
pip install --upgrade .

NOTE for developers: If you wish to fork/clone the respository and make changes to any of the ctlearn modules, the package must be reinstalled for the changes to take effect.

Dependencies

  • Python 3.6.5
  • TensorFlow 1.9.0
  • Keras
  • NumPy
  • Astropy
  • PyTables
  • PyYAML
  • SciPy
  • Libraries used only in plotting scripts (optional)
  • Matplotlib
  • Pillow
  • Scikit-learn

Uninstall LCC

Remove Anaconda Environment

First, remove the conda environment in which LCC is installed and all its dependencies:

conda remove --name lcc --all

Remove LCC

Next, completely remove LCC from your system:

rm -rf </installation/path>/LightCurveClassifier