Menu
Export and integrate MULLER datasets with other formats - Arrow, Parquet, JSON, NumPy, MindRecord. Use when user wants to export data, convert formats, or integrate with other frameworks.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Advanced query operations for MULLER datasets - indexing, vector search, aggregation, and complex filtering. Use when user wants to create indexes, perform vector similarity search, or run aggregation queries.
Create, query, and manage MULLER datasets (multimodal data lake with version control). Use when user wants to work with datasets, create tensors, append data, query samples, or inspect dataset information.
Git-like version control for MULLER datasets - commit, branch, merge, diff, and conflict resolution. Use when user wants to version datasets, create branches, merge changes, or view history.
| name | muller-export |
| description | Export and integrate MULLER datasets with other formats - Arrow, Parquet, JSON, NumPy, MindRecord. Use when user wants to export data, convert formats, or integrate with other frameworks. |
| compatibility | Requires Python 3.11+, muller package installed |
DO NOT create new Python files. Always use the existing script:
scripts/export.py for all export and integration operationsExecute the script directly with python3 command. Never write new scripts to the project root.
When executing operations:
Use this skill when the user wants to:
Handles export and format conversion operations.
Operations:
to-arrow - Export to Apache Arrow formatto-parquet - Export to Parquet filesto-json - Export to JSON formatto-numpy - Convert tensors to NumPy arraysto-mindrecord - Export to MindRecord formatget-info - Get export informationUsage:
# Export to Arrow
python3 .claude/skills/muller-export/scripts/export.py to-arrow \
--path ./my_dataset --output ./output.arrow
# Export to Parquet
python3 .claude/skills/muller-export/scripts/export.py to-parquet \
--path ./my_dataset --output ./output_dir
# Export to JSON
python3 .claude/skills/muller-export/scripts/export.py to-json \
--path ./my_dataset --output ./output.json
# Convert tensor to NumPy
python3 .claude/skills/muller-export/scripts/export.py to-numpy \
--path ./my_dataset --tensor embeddings --output ./embeddings.npy
# Export to MindRecord
python3 .claude/skills/muller-export/scripts/export.py to-mindrecord \
--path ./my_dataset --output ./output.mindrecord
Export formats: Arrow (columnar exchange), Parquet (storage/analytics), JSON (web APIs), NumPy (ML frameworks), MindRecord (MindSpore).
For detailed workflows, check references/ directory.