yangdong commited on
Commit
f80a979
1 Parent(s): 25cf951

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +2 -1
app.py CHANGED
@@ -15,9 +15,10 @@ def get_tokenizer(name):
15
  def examine(tokenizer, tok_name, inp):
16
  enc = tokenizer.encode(inp)
17
  reconstructed_inp = tokenizer.decode(enc)
 
18
  st.header(f"{tok_name}")
19
  st.subheader(f"#tokens = {len(enc)}")
20
- st.text(f"Exact match = {reconstructed_inp == inp}")
21
  st.code(reconstructed_inp)
22
  st.write(enc)
23
 
 
15
  def examine(tokenizer, tok_name, inp):
16
  enc = tokenizer.encode(inp)
17
  reconstructed_inp = tokenizer.decode(enc)
18
+ reconstructed_inp2 = tokenizer.decode(enc, skip_special_tokens=True)
19
  st.header(f"{tok_name}")
20
  st.subheader(f"#tokens = {len(enc)}")
21
+ st.text(f"Exact match = {reconstructed_inp2 == inp}")
22
  st.code(reconstructed_inp)
23
  st.write(enc)
24