Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -1,29 +1,25 @@
|
|
1 |
---
|
2 |
-
|
3 |
-
|
4 |
-
- name: prompt
|
5 |
-
dtype: string
|
6 |
-
- name: completion
|
7 |
-
dtype: string
|
8 |
-
splits:
|
9 |
-
- name: train
|
10 |
-
num_bytes: 181611007
|
11 |
-
num_examples: 116722
|
12 |
-
- name: validation
|
13 |
-
num_bytes: 10037679
|
14 |
-
num_examples: 6447
|
15 |
-
- name: test
|
16 |
-
num_bytes: 10217787
|
17 |
-
num_examples: 6553
|
18 |
-
download_size: 122702998
|
19 |
-
dataset_size: 201866473
|
20 |
-
configs:
|
21 |
-
- config_name: default
|
22 |
-
data_files:
|
23 |
-
- split: train
|
24 |
-
path: data/train-*
|
25 |
-
- split: validation
|
26 |
-
path: data/validation-*
|
27 |
-
- split: test
|
28 |
-
path: data/test-*
|
29 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
tags:
|
3 |
+
- trl
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
---
|
5 |
+
|
6 |
+
# TL;DR Dataset
|
7 |
+
|
8 |
+
## Summary
|
9 |
+
|
10 |
+
The TL;DR dataset is a processed version of Reddit posts, specifically curated to train models using the [TRL library](https://github.com/huggingface/trl) for summarization tasks. It leverages the common practice on Reddit where users append "TL;DR" (Too Long; Didn't Read) summaries to lengthy posts, providing a rich source of paired text data for training summarization models.
|
11 |
+
|
12 |
+
## Data Structure
|
13 |
+
|
14 |
+
- **Format**: [Standard](https://huggingface.co/docs/trl/main/dataset_formats#standard)
|
15 |
+
- **Type**: [Prompt-completion](https://huggingface.co/docs/trl/main/dataset_formats#prompt-completion)
|
16 |
+
|
17 |
+
Columns:
|
18 |
+
- `"pompt"`: The unabridged Reddit post.
|
19 |
+
- `"completion"`: The concise "TL;DR" summary appended by the author.
|
20 |
+
|
21 |
+
This structure enables models to learn the relationship between detailed content and its abbreviated form, enhancing their summarization capabilities.
|
22 |
+
|
23 |
+
## Generation script
|
24 |
+
|
25 |
+
The script used to generate this dataset can be found [here](https://github.com/huggingface/trl/blob/main/examples/datasets/tldr.py).
|