Spaces:
Running
Running
lastdefiance20
commited on
Commit
β’
bc0f422
1
Parent(s):
84481d9
Update some contant, and gradio version
Browse files- README.md +2 -2
- app.py +5 -4
- bottom_logo.png +0 -0
- content.py +20 -7
README.md
CHANGED
@@ -4,9 +4,9 @@ emoji: π
|
|
4 |
colorFrom: blue
|
5 |
colorTo: purple
|
6 |
sdk: gradio
|
7 |
-
sdk_version:
|
8 |
app_file: app.py
|
9 |
-
pinned:
|
10 |
license: apache-2.0
|
11 |
---
|
12 |
|
|
|
4 |
colorFrom: blue
|
5 |
colorTo: purple
|
6 |
sdk: gradio
|
7 |
+
sdk_version: 5.6.0
|
8 |
app_file: app.py
|
9 |
+
pinned: true
|
10 |
license: apache-2.0
|
11 |
---
|
12 |
|
app.py
CHANGED
@@ -23,11 +23,11 @@ with gr.Blocks(title=f'{benchname} Leaderboard', head=head_style) as demo:
|
|
23 |
gr.Markdown(intro_md)
|
24 |
|
25 |
with gr.Tabs(elem_classes='tab-buttons') as tabs:
|
26 |
-
with gr.TabItem('Main Leaderboard', elem_id='main', id=0):
|
27 |
table, type_map = build_df()
|
28 |
headers = [coln for coln in table.columns if not coln.startswith('_')]
|
29 |
datatypes = [type_map[x] for x in headers]
|
30 |
-
colwidths = [
|
31 |
|
32 |
with gr.Row():
|
33 |
model_name = gr.Textbox(
|
@@ -84,15 +84,16 @@ with gr.Blocks(title=f'{benchname} Leaderboard', head=head_style) as demo:
|
|
84 |
model_size.change(filter_df, [model_name, model_size, model_type], data_component)
|
85 |
model_type.change(filter_df, [model_name, model_size, model_type], data_component)
|
86 |
|
87 |
-
with gr.TabItem('About', elem_id='about', id=1):
|
88 |
gr.Markdown(about_md)
|
89 |
|
90 |
-
with gr.TabItem('Submit your model', elem_id='submit', id=2):
|
91 |
gr.Markdown(submit_md)
|
92 |
|
93 |
# with gr.Row():
|
94 |
# with gr.Accordion('Citation', open=False):
|
95 |
# gr.Markdown(citation_md)
|
|
|
96 |
|
97 |
|
98 |
if __name__=="__main__":
|
|
|
23 |
gr.Markdown(intro_md)
|
24 |
|
25 |
with gr.Tabs(elem_classes='tab-buttons') as tabs:
|
26 |
+
with gr.TabItem('π
Main Leaderboard', elem_id='main', id=0):
|
27 |
table, type_map = build_df()
|
28 |
headers = [coln for coln in table.columns if not coln.startswith('_')]
|
29 |
datatypes = [type_map[x] for x in headers]
|
30 |
+
colwidths = [80, 400, 120, 130] + [155 for _ in headers[4:]]
|
31 |
|
32 |
with gr.Row():
|
33 |
model_name = gr.Textbox(
|
|
|
84 |
model_size.change(filter_df, [model_name, model_size, model_type], data_component)
|
85 |
model_type.change(filter_df, [model_name, model_size, model_type], data_component)
|
86 |
|
87 |
+
with gr.TabItem('π About', elem_id='about', id=1):
|
88 |
gr.Markdown(about_md)
|
89 |
|
90 |
+
with gr.TabItem('π Submit your model', elem_id='submit', id=2):
|
91 |
gr.Markdown(submit_md)
|
92 |
|
93 |
# with gr.Row():
|
94 |
# with gr.Accordion('Citation', open=False):
|
95 |
# gr.Markdown(citation_md)
|
96 |
+
gr.Markdown(Bottom_logo)
|
97 |
|
98 |
|
99 |
if __name__=="__main__":
|
bottom_logo.png
ADDED
content.py
CHANGED
@@ -1,12 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
benchname = 'KOFFVQA'
|
2 |
|
|
|
|
|
3 |
intro_md = f'''
|
|
|
4 |
|
5 |
-
|
6 |
|
7 |
-
|
8 |
|
9 |
-
|
10 |
|
11 |
'''.strip()
|
12 |
|
@@ -14,13 +25,15 @@ about_md = f'''
|
|
14 |
|
15 |
# About
|
16 |
|
17 |
-
{benchname}
|
|
|
|
|
18 |
|
19 |
-
|
20 |
|
21 |
## News
|
22 |
|
23 |
-
* **2024-12-06**:
|
24 |
|
25 |
'''.strip()
|
26 |
|
@@ -28,6 +41,6 @@ submit_md = f'''
|
|
28 |
|
29 |
# Submit (coming soon)
|
30 |
|
31 |
-
|
32 |
|
33 |
'''.strip()
|
|
|
1 |
+
import os
|
2 |
+
import base64
|
3 |
+
|
4 |
+
current_dir = os.path.dirname(os.path.realpath(__file__))
|
5 |
+
|
6 |
+
with open(os.path.join(current_dir, "bottom_logo.png"), "rb") as image_file:
|
7 |
+
bottom_logo = base64.b64encode(image_file.read()).decode("utf-8")
|
8 |
+
|
9 |
benchname = 'KOFFVQA'
|
10 |
|
11 |
+
Bottom_logo = f'''<img src="data:image/jpeg;base64,{bottom_logo}" style="width:20%;display:block;margin-left:auto;margin-right:auto">'''
|
12 |
+
|
13 |
intro_md = f'''
|
14 |
+
# KOFFVQA Leaderboard
|
15 |
|
16 |
+
The {benchname}π is a Free-Form VQA benchmark dataset designed to evaluate Vision-Language Models (VLMs) in Korean language environments. Unlike traditional multiple-choice or predefined answer formats, KOFFVQA challenges models to generate open-ended, natural-language answers to visually grounded questions. This allows for a more comprehensive assessment of a model's ability to understand and generate nuanced Korean responses.
|
17 |
|
18 |
+
The dataset encompasses diverse real-world scenarios, including object attributes, recognition, relationship, etc.
|
19 |
|
20 |
+
The page will be continuously updated and will accept requests to add models to the leaderboard. For more details, please refer to the "Submit" tab.
|
21 |
|
22 |
'''.strip()
|
23 |
|
|
|
25 |
|
26 |
# About
|
27 |
|
28 |
+
The {benchname} benchmark is designed to evaluate and compare the performance of Vision-Language Models (VLMs) in Korean language environments.
|
29 |
+
|
30 |
+
This benchmark includes a total of 275 Korean questions across 10 tasks. The questions are open-ended, free-form VQA (Visual Question Answering) with objective answers, allowing responses without strict format constraints.
|
31 |
|
32 |
+
We will add more information about this benchmark soon.
|
33 |
|
34 |
## News
|
35 |
|
36 |
+
* **2024-12-06**: Leaderboard Release!
|
37 |
|
38 |
'''.strip()
|
39 |
|
|
|
41 |
|
42 |
# Submit (coming soon)
|
43 |
|
44 |
+
We are not accepting model addition requests at the moment. Once the request system is established, we will start accepting requests.
|
45 |
|
46 |
'''.strip()
|