Spaces:
Running
on
Zero
Running
on
Zero
Commit
•
3255b79
1
Parent(s):
7d6f3d4
Fixed image size and bullet lists
Browse files
app.py
CHANGED
@@ -14,17 +14,23 @@ Welcome to our Huggingface space, where you can explore the capabilities of TURN
|
|
14 |
- **Powerful Architecture:** TURNA contains 1.1B parameters, and was pre-trained with an encoder-decoder architecture following the UL2 framework on 43B tokens from various domains.
|
15 |
- **Diverse Training Data:** Our model is trained on a varied dataset of 43 billion tokens, covering a wide array of domains.
|
16 |
- **Broad Applications:** TURNA is fine-tuned for a variety of generation and understanding tasks, including:
|
17 |
-
|
18 |
- Summarization
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
|
|
|
|
|
|
|
|
27 |
|
|
|
|
|
|
|
28 |
Refer to our [paper](https://arxiv.org/abs/2401.14373) for more details.
|
29 |
|
30 |
### Citation
|
@@ -38,11 +44,6 @@ Refer to our [paper](https://arxiv.org/abs/2401.14373) for more details.
|
|
38 |
primaryClass={cs.CL}
|
39 |
}
|
40 |
```
|
41 |
-
|
42 |
-
**Note:** First inference might take time as the models are downloaded on-the-go.
|
43 |
-
|
44 |
-
*TURNA can generate toxic content or provide erroneous information. Double-check before usage.*
|
45 |
-
|
46 |
"""
|
47 |
|
48 |
|
@@ -138,12 +139,10 @@ def turna(input, max_new_tokens, length_penalty,
|
|
138 |
|
139 |
with gr.Blocks(theme="abidlabs/Lime") as demo:
|
140 |
gr.Markdown("# TURNA")
|
141 |
-
gr.
|
|
|
142 |
|
143 |
gr.Markdown(DESCRIPTION)
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
|
148 |
with gr.Tab("Sentiment Analysis"):
|
149 |
gr.Markdown("TURNA fine-tuned on sentiment analysis. Enter text to analyse sentiment and pick the model (tweets or product reviews).")
|
@@ -265,4 +264,7 @@ with gr.Blocks(theme="abidlabs/Lime") as demo:
|
|
265 |
|
266 |
sum_submit.click(summarize, inputs=[sum_input, sum_choice], outputs=sum_output)
|
267 |
sum_examples = gr.Examples(examples = long_text, inputs = [sum_input, sum_choice], outputs=sum_output, fn=summarize)
|
|
|
|
|
|
|
268 |
demo.launch()
|
|
|
14 |
- **Powerful Architecture:** TURNA contains 1.1B parameters, and was pre-trained with an encoder-decoder architecture following the UL2 framework on 43B tokens from various domains.
|
15 |
- **Diverse Training Data:** Our model is trained on a varied dataset of 43 billion tokens, covering a wide array of domains.
|
16 |
- **Broad Applications:** TURNA is fine-tuned for a variety of generation and understanding tasks, including:
|
|
|
17 |
- Summarization
|
18 |
+
- Paraphrasing
|
19 |
+
- News title generation
|
20 |
+
- Sentiment classification
|
21 |
+
- Text categorization
|
22 |
+
- Named entity recognition
|
23 |
+
- Part-of-speech tagging
|
24 |
+
- Semantic textual similarity
|
25 |
+
- Natural language inference
|
26 |
+
|
27 |
+
**Note:** First inference might take time as the models are downloaded on-the-go.
|
28 |
+
|
29 |
+
*TURNA can generate toxic content or provide erroneous information. Double-check before usage.*
|
30 |
|
31 |
+
"""
|
32 |
+
|
33 |
+
CITATION = """
|
34 |
Refer to our [paper](https://arxiv.org/abs/2401.14373) for more details.
|
35 |
|
36 |
### Citation
|
|
|
44 |
primaryClass={cs.CL}
|
45 |
}
|
46 |
```
|
|
|
|
|
|
|
|
|
|
|
47 |
"""
|
48 |
|
49 |
|
|
|
139 |
|
140 |
with gr.Blocks(theme="abidlabs/Lime") as demo:
|
141 |
gr.Markdown("# TURNA")
|
142 |
+
with gr.Row():
|
143 |
+
gr.Image("images/turna-logo.png", width=100)
|
144 |
|
145 |
gr.Markdown(DESCRIPTION)
|
|
|
|
|
|
|
146 |
|
147 |
with gr.Tab("Sentiment Analysis"):
|
148 |
gr.Markdown("TURNA fine-tuned on sentiment analysis. Enter text to analyse sentiment and pick the model (tweets or product reviews).")
|
|
|
264 |
|
265 |
sum_submit.click(summarize, inputs=[sum_input, sum_choice], outputs=sum_output)
|
266 |
sum_examples = gr.Examples(examples = long_text, inputs = [sum_input, sum_choice], outputs=sum_output, fn=summarize)
|
267 |
+
|
268 |
+
gr.Markdown(CITATION)
|
269 |
+
|
270 |
demo.launch()
|