| --- |
| extra_gated_heading: 'Before requesting access' |
| extra_gated_description: 'Like this dataset and follow EdgeArbiter, then request access. Requests are checked about once a minute while the approval laptop is online; otherwise they remain pending.' |
| configs: |
| - config_name: XIV |
| data_files: |
| - split: train |
| path: data/XIV_r0001.parquet |
| --- |
| # RKBI_Data |
| |
| ## Access |
| |
| 1. Like this dataset using the heart button. |
| 2. Follow [@EdgeArbiter](https://huggingface.co/EdgeArbiter). |
| 3. Log in and request access. |
| |
| Access is approved automatically after both conditions are verified. Requests |
| missing either condition stay pending. Checks run about once a minute while |
| the owner's laptop is awake, signed in, and online; approval may take longer |
| when it is unavailable. Approved downloads require your own Hugging Face token. |
| |
| |
| One daily tabular subset identified as **XIV**, containing one current full-history |
| Parquet file. Instrument names use reversed-character aliases. |
| |
| Descriptions, reversible column mappings, provenance, original schema metadata, |
| coverage, missing-value counts, and validated checksums are in |
| [metadata.enc](metadata.enc), encrypted with the owner's existing key. |
| |
| The records are unencrypted and available to approved users. Metadata encryption protects descriptions; |
| published values and aliases can still reveal their meaning. Updates are manual |
| after validation; no real-time delivery or fixed schedule is promised. |
| |
| First log in locally with `hf auth login` after your request is approved. |
| |
| ```python |
| from datasets import load_dataset |
|
|
| series = load_dataset("EdgeArbiter/RKBI_Data", "XIV", split="train", token=True) |
| ``` |
| |
