latent-space-builder 0.0.1

A tool for building latent spaces.

LatentSpaceBuilder

The Latent Space Builder builds the latent space for an image dataset.

Getting started

Clone this repository.

git clone https://github.com/compSPI/LatentSpaceBuilder.git

Change the current working directory to the root of this repository.

cd LatentSpaceBuilder

Download from the Anaconda Cloud and install the Python environment that has the dependencies required to run the code.

conda env create compSPI/compSPI

Activate the environment.

conda activate compSPI

Install the kernel.

python -m ipykernel install --user --name compSPI --display-name "Python (compSPI)"

Exit the environment.

conda deactivate

Running the notebook

Run jupyter notebook.

jupyter notebook 

Open the tutorial notebook latent_space_builder.ipynb.

Change the Python kernel to compSPI.

Set dataset_file to an HDF5 file containing the dataset.

dataset_file = '../data/cspi_synthetic_dataset_diffraction_patterns_1024x1040.hdf5'

Run the notebook.

Installation

The project package can be installed by running the following command.

python setup.py install

Code structure

The relevant files and folders in this repository are described below:

• README.md: Highlights the usefulness of the Latent Space builder.

• latent_space_builder.ipynb: Provides a tutorial notebook for using the Latent Space Builder.

• latent_space_builder/: Contains the Python file required to run the notebook.