This directory contains files generated by sampling point clouds from meshes in the EuLearn dataset. While these samples can also be produced using the code provided in the Architectures directory, we include precomputed versions in the sampled folder (in compressed format) to save users time, as the generation process can be time-consuming.
To reconstruct the full tar.xz archive (sampling.tar.gz) from the split files in the sampled folder, use the following command:
cat sampling_part_* > sampling.tar.xz
Once assembled, you can extract the contents with:
tar -xvf sampling.tar.xz
This will produce a directory containing 11 .tar.gz files, each corresponding to sampled point clouds grouped and compressed (tar.gz) by genus.
Remark: Before extracting each .tar.gz file, ensure your machine has sufficient free space to store the uncompressed samples. To this end, consider the following example to estimate the extracted size:
gzip -l sampled_g0_complete.tar.gz