Update README.md
Browse files
README.md
CHANGED
|
@@ -6,7 +6,9 @@ configs:
|
|
| 6 |
- config_name: default
|
| 7 |
data_files:
|
| 8 |
- split: train
|
| 9 |
-
path:
|
|
|
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
Here are the _mostly_ original files for some of the forums I scraped in the past, repacked as HTML strings + some metadata on a **per-thread** basis instead of a per-message basis, which should make them more convenient to handle. Unlike [the other archive](https://huggingface.co/datasets/lemonilia/Roleplay-Forums_2023-04), they shouldn't have issues with spaces between adjacent HTML tags, as that occurred by mistake in an intermediate step where single messages were extracted from the pages. Unfortunately, I do not have anymore the original files for most of the forums scraped in 2023; those would need to be scraped again.
|
|
|
|
| 6 |
- config_name: default
|
| 7 |
data_files:
|
| 8 |
- split: train
|
| 9 |
+
path: '*.parquet'
|
| 10 |
+
size_categories:
|
| 11 |
+
- 100K<n<1M
|
| 12 |
---
|
| 13 |
|
| 14 |
Here are the _mostly_ original files for some of the forums I scraped in the past, repacked as HTML strings + some metadata on a **per-thread** basis instead of a per-message basis, which should make them more convenient to handle. Unlike [the other archive](https://huggingface.co/datasets/lemonilia/Roleplay-Forums_2023-04), they shouldn't have issues with spaces between adjacent HTML tags, as that occurred by mistake in an intermediate step where single messages were extracted from the pages. Unfortunately, I do not have anymore the original files for most of the forums scraped in 2023; those would need to be scraped again.
|