Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import AutoTokenizer
|
3 |
+
from huggingface_hub import HfApi
|
4 |
+
from gradio_huggingfacehub_search import HuggingfaceHubSearch
|
5 |
+
|
6 |
+
def count_tokens(model_id, text):
|
7 |
+
try:
|
8 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
9 |
+
tokens = tokenizer.encode(text)
|
10 |
+
|
11 |
+
token_count = len(tokens)
|
12 |
+
|
13 |
+
return f"Number of tokens: {token_count}"
|
14 |
+
except Exception as e:
|
15 |
+
return f"Error: {str(e)}"
|
16 |
+
|
17 |
+
with gr.Blocks() as iface:
|
18 |
+
gr.Markdown("# Universal Tokenizer - Token Counter")
|
19 |
+
gr.Markdown("This app counts the number of tokens in the provided text using any tokenizer from a Hugging Face model.")
|
20 |
+
|
21 |
+
model_id = HuggingfaceHubSearch(
|
22 |
+
label="Select a model repo with a tokenizer",
|
23 |
+
placeholder="Search for a model on Hugging Face",
|
24 |
+
search_type="model",
|
25 |
+
)
|
26 |
+
|
27 |
+
text_input = gr.Textbox(lines=5, placeholder="Enter your text here...")
|
28 |
+
|
29 |
+
output = gr.Textbox(label="Result")
|
30 |
+
|
31 |
+
btn = gr.Button("Count Tokens")
|
32 |
+
btn.click(fn=count_tokens, inputs=[model_id, text_input], outputs=output)
|
33 |
+
|
34 |
+
iface.launch()
|