VTSNLP commited on
Commit
dea46ed
·
verified ·
1 Parent(s): dccb75b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -0
README.md CHANGED
@@ -45,3 +45,17 @@ We use [NeMo Curator](https://github.com/NVIDIA/NeMo-Curator) to curate the coll
45
  3. Quality Filtering:
46
  4. Heuristic Filtering: Applies rules-based filters to remove low-quality content.
47
  5. Classifier-Based Filtering: Uses machine learning to classify and filter documents based on quality.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45
  3. Quality Filtering:
46
  4. Heuristic Filtering: Applies rules-based filters to remove low-quality content.
47
  5. Classifier-Based Filtering: Uses machine learning to classify and filter documents based on quality.
48
+
49
+ #### Dataset Statistics
50
+
51
+ ** Content diversity
52
+ ![Domain proportion in curated dataset](https://cdn-uploads.huggingface.co/production/uploads/661766c00c68b375f3f0ccc3/mW6Pct3uyP_XDdGmE8EP3.png)
53
+
54
+ ** Character based metrics
55
+ ![Box plots of percentage of symbols, numbers, and whitespace characters compared to the total characters, word counts and average word lengths](https://cdn-uploads.huggingface.co/production/uploads/661766c00c68b375f3f0ccc3/W9TQjM2vcC7uXozyERHSQ.png)
56
+
57
+ ** Token count distribution
58
+ ![Distribution of document sizes (in terms of token count)](https://cdn-uploads.huggingface.co/production/uploads/661766c00c68b375f3f0ccc3/PDelYpBI0DefSmQgFONgE.png)
59
+
60
+ ** Embedding visualization
61
+ ![UMAP visualization of 5% of the dataset](https://cdn-uploads.huggingface.co/production/uploads/661766c00c68b375f3f0ccc3/sfeoZWuQ7DcSpbmUOJ12r.png)