From c47f283ff7d394316baa829b85393738fa9c20af Mon Sep 17 00:00:00 2001 From: EvaBardou <bardou@teklia.com> Date: Mon, 13 Nov 2023 12:07:11 +0100 Subject: [PATCH] Update dataset archive extension to .tar.zst --- worker_generic_training_dataset/worker.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/worker_generic_training_dataset/worker.py b/worker_generic_training_dataset/worker.py index 43109d8..caef383 100644 --- a/worker_generic_training_dataset/worker.py +++ b/worker_generic_training_dataset/worker.py @@ -355,11 +355,11 @@ class DatasetExtractor(DatasetWorker): casted_elements = list(map(_format_element, elements)) self.process_split(split_name, casted_elements) - # TAR + ZSTD the cache and the images folder, and store as task artifact - zstd_archive_path: Path = self.work_dir / f"{dataset.id}.zstd" - logger.info(f"Compressing the images to {zstd_archive_path}") + # TAR + ZST the cache and the images folder, and store as task artifact + zst_archive_path: Path = self.work_dir / f"{dataset.id}.tar.zst" + logger.info(f"Compressing the images to {zst_archive_path}") create_tar_zst_archive( - source=self.data_folder_path, destination=zstd_archive_path + source=self.data_folder_path, destination=zst_archive_path ) self.data_folder.cleanup() -- GitLab