Datasets:
Document UDPipe Bosque enrichment notebook
Browse files- source_code/README.md +7 -0
source_code/README.md
CHANGED
|
@@ -36,9 +36,16 @@ modeling/NER stages.
|
|
| 36 |
Contains exploratory predictive modeling and interpretability experiments
|
| 37 |
over note helpfulness/consensus.
|
| 38 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
## Notes
|
| 40 |
|
| 41 |
- Tweet text is not redistributed. Only tweet identifiers appear in the dataset.
|
| 42 |
- The notebooks are intended as transparent research code, not as a packaged
|
| 43 |
one-command production pipeline.
|
| 44 |
- The released Parquet files are the authoritative dataset artifacts.
|
|
|
|
|
|
|
|
|
| 36 |
Contains exploratory predictive modeling and interpretability experiments
|
| 37 |
over note helpfulness/consensus.
|
| 38 |
|
| 39 |
+
7. `07_udpipe_bosque_enrichment_colab.ipynb`
|
| 40 |
+
Optional Google Colab notebook that creates an automatic Universal
|
| 41 |
+
Dependencies syntax layer with UDPipe Bosque and uploads it to
|
| 42 |
+
`syntax_ud_bosque/` in the existing Hugging Face dataset repository.
|
| 43 |
+
|
| 44 |
## Notes
|
| 45 |
|
| 46 |
- Tweet text is not redistributed. Only tweet identifiers appear in the dataset.
|
| 47 |
- The notebooks are intended as transparent research code, not as a packaged
|
| 48 |
one-command production pipeline.
|
| 49 |
- The released Parquet files are the authoritative dataset artifacts.
|
| 50 |
+
- Optional derived layers generated by external models may have separate
|
| 51 |
+
license metadata from the CC0 core dataset.
|