Create README.md
#3
by
FalconNet
- opened
README.md
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# π Dataset Overview
|
2 |
+
|
3 |
+
The **drlc-leaderboard-data** dataset is part of the Hugging Face **projects** collection. It is primarily a **tabular** dataset with a focus on **text** modalities. The dataset is available in **CSV** format and contains between **10K - 100K** entries.
|
4 |
+
|
5 |
+
# π Dataset Details
|
6 |
+
|
7 |
+
The dataset comprises a single split named *train*, containing **32.4K rows**. It includes various columns, each serving a distinct purpose. The **Ranking** column is of type `int64`, while the **User** and **Model** columns are strings with varying lengths. The **Results**, **Mean Reward**, and **Std Reward** columns are of type `float64`.
|
8 |
+
|
9 |
+
# π Data Visualization
|
10 |
+
|
11 |
+
The dataset viewer provides visual insights into the distribution of data across different columns. For instance, the **Ranking** column displays a bar chart with a peak at **10.1K**. The **User** and **Model** columns show diverse entries, with the latter having a length of **69**.
|
12 |
+
|
13 |
+
# π Key Metrics
|
14 |
+
|
15 |
+
- **Results**: Ranges from **-6,040** to **200K**.
|
16 |
+
- **Mean Reward**: Spans from **-4,851** to **407K**.
|
17 |
+
- **Std Reward**: Values range from **-1,000** to **320K**.
|
18 |
+
|