Add ABC_Dataset_abci_pkl_json_aug.tar and scripts for v0.4 (pitch-shift augmentation, exclude xml)

ABC_Dataset_abci_pkl_json_aug.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03ffeea17dcb3b52acc83e46918c15cf8c185f7d5f63edf536fc47d52d2e9cf7
+size 5539852326

README.md

@@ -2,6 +2,25 @@
 license: unknown
 dataset: mimbres/text2score
 ---
+
+#### 🚀 Download & Install Dataset 
+```
+pip install -U huggingface_hub hf-transfer
+export HF_HUB_ENABLE_HF_TRANSFER=1
+huggingface-cli download mimbres/text2score --local-dir ABC_Dataset
+cd ABC_Dataset
+tar xf ABC_Dataset_abci_pkl_json_aug.tar
+```
+
+#### v0.4
+- new plan json files
+- pitch-shift augmentation 
+- exclude xml files
+
+#### v0.3
+- new plan json files
+- full data including ASAP
+
 #### v0.2
 - The key information in the ABC ...
 - Changing the file extension fr ...

extract_tar.sh

@@ -0,0 +1,63 @@
+#!/bin/bash
+
+# =================================================================
+# Configuration
+# Modify these variables to match your environment.
+# =================================================================
+
+# Prefix for the split files (e.g., abc_dataset_part_aa, abc_dataset_part_ab, etc.)
+FILE_PREFIX="abc_dataset_part_"
+
+# Name of the final merged .tar file
+MERGED_TAR_FILE="ABC_Dataset_full.tar"
+
+# Directory to extract files into
+EXTRACT_DIR="ABC_Dataset_extracted"
+
+
+# =================================================================
+# Script Body
+# Do not modify below this line unless you know what you are doing.
+# =================================================================
+
+# Exit immediately if a command exits with a non-zero status.
+set -e
+
+echo ">> 1. Verifying split files exist..."
+files=(${FILE_PREFIX}*)
+if [ ! -f "${files[0]}" ]; then
+    echo "Error: Could not find split files starting with '${FILE_PREFIX}'."
+    echo "Ensure this script is in the same directory as the split files and FILE_PREFIX is correct."
+    exit 1
+fi
+echo "Split files found. Starting merge."
+
+echo ""
+echo ">> 2. Merging split files..."
+# The wildcard (*) is automatically sorted alphabetically by the shell (aa, ab, ac...).
+cat ${FILE_PREFIX}* > ${MERGED_TAR_FILE}
+echo "Merge complete. Output file: ${MERGED_TAR_FILE}"
+
+echo ""
+echo ">> 3. Creating destination directory..."
+mkdir -p ${EXTRACT_DIR}
+echo "Directory ready: ${EXTRACT_DIR}"
+
+echo ""
+echo ">> 4. Extracting TAR archive..."
+# -x: extract, -v: verbose, -f: file, -C: change to directory
+tar -xvf ${MERGED_TAR_FILE} -C ${EXTRACT_DIR}
+echo "Extraction complete."
+
+echo ""
+echo "=========================================="
+echo "🎉 All tasks completed successfully!"
+echo "Data is available in the '${EXTRACT_DIR}' directory."
+echo "=========================================="
+
+# Optional: Uncomment the lines below to delete the original split parts and the merged .tar file after completion.
+# echo ""
+# echo ">> 5. Cleaning up source files..."
+# rm ${FILE_PREFIX}*
+# rm ${MERGED_TAR_FILE}
+# echo "Cleanup complete."

pack_to_tar.sh

@@ -0,0 +1 @@
+cd ABC_Dataset && rsync -a --include='*/' --include='*.abci' --include='*.pkl' --include='*.json' --exclude='*' --no-inc-recursive . ../tmp_filtered && cd ../tmp_filtered && tar -cf ../ABC_Dataset_only_abci_pkl_json.tar --use-compress-program=pigz * && cd .. && rm -rf tmp_filtered

split_tar.sh

@@ -0,0 +1 @@
+split -b 40G -d ABC_Dataset_full.tar abc_dataset_part_

upload.sh

@@ -0,0 +1 @@
+huggingface-cli upload mimbres/text2score . . --repo-type dataset --commit-message "Add ABC_Dataset_abci_pkl_json_aug.tar and scripts for v0.4 (pitch-shift augmentation, exclude xml)"