cakiki commited on
Commit
392cbe8
β€’
1 Parent(s): 4f1840e

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +1 -1
app.py CHANGED
@@ -56,7 +56,7 @@ with gr.Blocks(css="#b {min-width:15px;background:transparent;}") as demo: #bord
56
  with gr.Column(scale=1):
57
  pass
58
  with gr.Column(scale=15):
59
- gr.Markdown("""<div style="text-align: justify"> This search tool was used to validate tokenization scheme for code retrieval for the BigCode project. We indexed the <a href="https://huggingface.co/bigcode/santacoder">πŸŽ… Santacoder</a> training dataset (Python, Java, and JavaScript) and use a (2,4)-gram tokenizer to build the index. This is the same tokenization scheme that ended up being used to power the <a href="https://huggingface.co/spaces/bigcode/search">⭐ StarCoder search tool</a>.</div>""")
60
  with gr.Column(scale=1):
61
  pass
62
  with gr.Row():
 
56
  with gr.Column(scale=1):
57
  pass
58
  with gr.Column(scale=15):
59
+ gr.Markdown("""<div style="text-align: justify"> This search tool was used to validate tokenization scheme for code retrieval for the BigCode project. We indexed the πŸŽ… <a href="https://huggingface.co/bigcode/santacoder">Santacoder</a> training dataset (Python, Java, and JavaScript) and use a (2,4)-gram tokenizer to build the index. This is the same tokenization scheme that ended up being used to power the ⭐ <a href="https://huggingface.co/spaces/bigcode/search">StarCoder search tool</a>.</div>""")
60
  with gr.Column(scale=1):
61
  pass
62
  with gr.Row():