Update app.py
Browse files
app.py
CHANGED
@@ -1,23 +1,74 @@
|
|
1 |
import gradio as gr
|
2 |
from transformers import pipeline
|
|
|
3 |
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
-
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
-
|
13 |
-
|
|
|
14 |
inputs=[
|
15 |
-
gr.Textbox(
|
16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
],
|
18 |
-
outputs=gr.Label(label="Classification
|
19 |
-
title="Zero-Shot Classification",
|
20 |
-
description="Classify
|
|
|
|
|
|
|
|
|
|
|
21 |
)
|
22 |
|
23 |
-
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
from transformers import pipeline
|
3 |
+
import torch
|
4 |
|
5 |
+
# Initialize the zero-shot classification pipeline
|
6 |
+
try:
|
7 |
+
classifier = pipeline(
|
8 |
+
"zero-shot-classification",
|
9 |
+
model="models/tasksource/ModernBERT-nli",
|
10 |
+
device=0 if torch.cuda.is_available() else -1
|
11 |
+
)
|
12 |
+
except Exception as e:
|
13 |
+
print(f"Error loading model: {e}")
|
14 |
+
classifier = None
|
15 |
|
16 |
+
def classify_text(text, candidate_labels):
|
17 |
+
"""
|
18 |
+
Perform zero-shot classification on input text.
|
19 |
+
|
20 |
+
Args:
|
21 |
+
text (str): Input text to classify
|
22 |
+
candidate_labels (str): Comma-separated string of possible labels
|
23 |
+
|
24 |
+
Returns:
|
25 |
+
dict: Dictionary containing labels and their corresponding scores
|
26 |
+
"""
|
27 |
+
if classifier is None:
|
28 |
+
return {"Error": "Model failed to load"}
|
29 |
+
|
30 |
+
try:
|
31 |
+
# Convert comma-separated string to list
|
32 |
+
labels = [label.strip() for label in candidate_labels.split(",")]
|
33 |
+
|
34 |
+
# Perform classification
|
35 |
+
result = classifier(text, labels)
|
36 |
+
|
37 |
+
# Create formatted output
|
38 |
+
output = {}
|
39 |
+
for label, score in zip(result["labels"], result["scores"]):
|
40 |
+
output[label] = f"{score:.4f}"
|
41 |
+
|
42 |
+
return output
|
43 |
+
|
44 |
+
except Exception as e:
|
45 |
+
return {"Error": str(e)}
|
46 |
|
47 |
+
# Create Gradio interface
|
48 |
+
iface = gr.Interface(
|
49 |
+
fn=classify_text,
|
50 |
inputs=[
|
51 |
+
gr.Textbox(
|
52 |
+
label="Text to classify",
|
53 |
+
placeholder="Enter text here...",
|
54 |
+
value="all cats are blue"
|
55 |
+
),
|
56 |
+
gr.Textbox(
|
57 |
+
label="Possible labels (comma-separated)",
|
58 |
+
placeholder="Enter labels...",
|
59 |
+
value="true,false"
|
60 |
+
)
|
61 |
],
|
62 |
+
outputs=gr.Label(label="Classification Results"),
|
63 |
+
title="Zero-Shot Text Classification",
|
64 |
+
description="Classify text into given categories without any training examples.",
|
65 |
+
examples=[
|
66 |
+
["all cats are blue", "true,false"],
|
67 |
+
["the sky is above us", "true,false"],
|
68 |
+
["birds can fly", "true,false,unknown"]
|
69 |
+
]
|
70 |
)
|
71 |
|
72 |
+
# Launch the app
|
73 |
+
if __name__ == "__main__":
|
74 |
+
iface.launch(share=True)
|