Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,7 +1,45 @@
|
|
1 |
import gradio as gr
|
|
|
2 |
|
3 |
-
def
|
4 |
-
|
|
|
|
|
5 |
|
6 |
-
|
7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
from transformers import AutoTokenizer
|
3 |
|
4 |
+
def load_tokenizer(repo_path):
|
5 |
+
try:
|
6 |
+
# Load the tokenizer from the provided repository path
|
7 |
+
tokenizer = AutoTokenizer.from_pretrained(repo_path, trust_remote_code=True)
|
8 |
|
9 |
+
messages = [
|
10 |
+
{"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
|
11 |
+
{"role": "user", "content": "Who are you?"},
|
12 |
+
]
|
13 |
+
|
14 |
+
input_ids = tokenizer.apply_chat_template(
|
15 |
+
messages,
|
16 |
+
add_generation_prompt=True,
|
17 |
+
tokenize=False,
|
18 |
+
)
|
19 |
+
|
20 |
+
# Extract relevant details about the tokenizer and chat template
|
21 |
+
details = {
|
22 |
+
"Tokenizer Name": tokenizer.name_or_path,
|
23 |
+
"Vocabulary Size": tokenizer.vocab_size,
|
24 |
+
"Model Max Length": tokenizer.model_max_length,
|
25 |
+
"Special Tokens": tokenizer.all_special_tokens,
|
26 |
+
"Chat Template": input_ids,
|
27 |
+
}
|
28 |
+
|
29 |
+
# Convert details to a formatted string for display
|
30 |
+
details_str = "\n".join([f"{key}: {value}" for key, value in details.items()])
|
31 |
+
return details_str
|
32 |
+
except Exception as e:
|
33 |
+
return str(e)
|
34 |
+
|
35 |
+
# Create the Gradio interface
|
36 |
+
iface = gr.Interface(
|
37 |
+
fn=load_tokenizer,
|
38 |
+
inputs=gr.Textbox(label="Hugging Face Repository Path (e.g., user/repo)"),
|
39 |
+
outputs=gr.Textbox(label="Tokenizer Details"),
|
40 |
+
title="Hugging Face Tokenizer Loader",
|
41 |
+
description="Enter the Hugging Face repository path to load the tokenizer and view its details."
|
42 |
+
)
|
43 |
+
|
44 |
+
# Launch the app
|
45 |
+
iface.launch()
|