cakiki commited on
Commit
5131ece
·
1 Parent(s): 19e7cae

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +14 -0
app.py ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import streamlit as st
2
+ from tokenizers.tools import EncodingVisualizer
3
+ from transformers import AutoTokenizer
4
+
5
+ st.set_page_config(page_title="BigCode Tokenizer", page_icon='👩‍💻', layout="wide")
6
+
7
+ tokenizer = AutoTokenizer.from_pretrained('bigcode/tokenizer', subfolder="digit-punctuation-bytelevel-bpe-py-js-java-50k")
8
+ visualizer = EncodingVisualizer(tokenizer=tokenizer._tokenizer, default_to_notebook=False)
9
+
10
+ text = st.text_area(label="", placeholder="Text to tokenize")
11
+ button_clicked = st.button("Tokenize")
12
+ if text or button_clicked:
13
+ st.write(f"The input was split into {len(tokenizer.tokenize(text))} tokens.")
14
+ st.components.v1.html(visualizer(text), height=1500)