Added Reference for dataset generation
Browse files
README.md
CHANGED
|
@@ -24,4 +24,12 @@ language:
|
|
| 24 |
pretty_name: 'wikipedia industrial technical '
|
| 25 |
size_categories:
|
| 26 |
- 100K<n<1M
|
| 27 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
pretty_name: 'wikipedia industrial technical '
|
| 25 |
size_categories:
|
| 26 |
- 100K<n<1M
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
This dataset was generated by filtering a subset from the wikipedia dataset -> "wikimedia/wikipedia" -> " 20231101.en"
|
| 32 |
+
|
| 33 |
+
Detailed information on how this was accomplished is given in this notebook. https://github.com/Umar-Azam/embedding_finetuner_wiki/tree/main
|
| 34 |
+
|
| 35 |
+
Short Explanation : We have a list of keywords to check against. Each wikipedia text is tokenized into word sets and the "hits" value contains the number of our filter keywords present in each text. Only the items with >4 matches are then filtered to generate this dataset.
|