data_factory_20k

# data_factory_20k bundle This dataset repo stores the train-only release as split tar.gz parts. - Source release directory: data_factory_20k_train_only_20260415_1935 - Logical archive name: data_factory_20k_train_only_20260415_1935.tar.gz - Part count: 36 - Split size: 5368709120 bytes Download every `*.part-*` file in this repo, then reconstruct and extract the archive. ## Restore ```bash cat data_factory_20k_train_only_20260415_1935.tar.gz.part-* > data_factory_20k_train_only_20260415_1935.tar.gz tar -xzf data_factory_20k_train_only_20260415_1935.tar.gz ``` The extracted directory will be: - `data_factory_20k_train_only_20260415_1935/` ## Integrity - See `bundle_manifest.json` for per-part SHA256 values and the combined archive SHA256.