Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -144,7 +144,6 @@ def evaluate_predictions(prediction_file, model_name, add_to_leaderboard):
|
|
| 144 |
|
| 145 |
initialize_leaderboard_file()
|
| 146 |
|
| 147 |
-
|
| 148 |
# Function to set default mode
|
| 149 |
css_tech_theme = """
|
| 150 |
body {
|
|
@@ -186,6 +185,19 @@ button:hover {
|
|
| 186 |
box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1);
|
| 187 |
}
|
| 188 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 189 |
.dataframe {
|
| 190 |
color: #333333;
|
| 191 |
background-color: #ffffff;
|
|
@@ -202,17 +214,17 @@ with gr.Blocks(css=css_tech_theme) as demo:
|
|
| 202 |
gr.Markdown("""
|
| 203 |
# π Mobile-MMLU Benchmark Competition
|
| 204 |
### π Welcome to the Competition Overview
|
| 205 |
-

|
| 210 |
|
| 211 |
-
with gr.Tabs():
|
| 212 |
-
with gr.TabItem("π Overview"):
|
| 213 |
gr.Markdown("""
|
| 214 |
## Overview
|
| 215 |
-
Welcome to the Mobile-MMLU Benchmark Competition
|
| 216 |
---
|
| 217 |
### What is Mobile-MMLU?
|
| 218 |
Mobile-MMLU is a benchmark designed to test the capabilities of LLMs optimized for mobile use. Contribute to advancing mobile AI systems by competing to achieve the highest accuracy.
|
|
@@ -245,7 +257,7 @@ For support, email: [Insert Email Address]
|
|
| 245 |
---
|
| 246 |
""")
|
| 247 |
|
| 248 |
-
with gr.TabItem("π€ Submission"):
|
| 249 |
with gr.Row():
|
| 250 |
file_input = gr.File(label="π Upload Prediction CSV", file_types=[".csv"], interactive=True)
|
| 251 |
model_name_input = gr.Textbox(label="ποΈ Model Name", placeholder="Enter your model name")
|
|
@@ -263,7 +275,7 @@ For support, email: [Insert Email Address]
|
|
| 263 |
outputs=[eval_status, overall_accuracy_display],
|
| 264 |
)
|
| 265 |
|
| 266 |
-
with gr.TabItem("π
Leaderboard"):
|
| 267 |
leaderboard_table = gr.Dataframe(
|
| 268 |
value=load_leaderboard(),
|
| 269 |
label="Leaderboard",
|
|
@@ -280,5 +292,3 @@ For support, email: [Insert Email Address]
|
|
| 280 |
gr.Markdown(f"**Last updated:** {LAST_UPDATED}")
|
| 281 |
|
| 282 |
demo.launch()
|
| 283 |
-
|
| 284 |
-
|
|
|
|
| 144 |
|
| 145 |
initialize_leaderboard_file()
|
| 146 |
|
|
|
|
| 147 |
# Function to set default mode
|
| 148 |
css_tech_theme = """
|
| 149 |
body {
|
|
|
|
| 185 |
box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1);
|
| 186 |
}
|
| 187 |
|
| 188 |
+
.tabs {
|
| 189 |
+
margin-bottom: 15px;
|
| 190 |
+
gap: 10px;
|
| 191 |
+
}
|
| 192 |
+
|
| 193 |
+
.tab-item {
|
| 194 |
+
background-color: #ece2f4;
|
| 195 |
+
border-radius: 6px;
|
| 196 |
+
padding: 10px;
|
| 197 |
+
box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1);
|
| 198 |
+
margin: 5px;
|
| 199 |
+
}
|
| 200 |
+
|
| 201 |
.dataframe {
|
| 202 |
color: #333333;
|
| 203 |
background-color: #ffffff;
|
|
|
|
| 214 |
gr.Markdown("""
|
| 215 |
# π Mobile-MMLU Benchmark Competition
|
| 216 |
### π Welcome to the Competition Overview
|
| 217 |
+

|
| 218 |
---
|
| 219 |
+
Welcome to the **Mobile-MMLU Benchmark Competition**. Here you can submit your predictions, view the leaderboard, and track your performance.
|
| 220 |
---
|
| 221 |
""")
|
| 222 |
|
| 223 |
+
with gr.Tabs(elem_id="tabs"):
|
| 224 |
+
with gr.TabItem("π Overview", elem_classes=["tab-item"]):
|
| 225 |
gr.Markdown("""
|
| 226 |
## Overview
|
| 227 |
+
Welcome to the **Mobile-MMLU Benchmark Competition**! Evaluate mobile-compatible Large Language Models (LLMs) on **16,186 scenario-based and factual questions** across **80 fields**.
|
| 228 |
---
|
| 229 |
### What is Mobile-MMLU?
|
| 230 |
Mobile-MMLU is a benchmark designed to test the capabilities of LLMs optimized for mobile use. Contribute to advancing mobile AI systems by competing to achieve the highest accuracy.
|
|
|
|
| 257 |
---
|
| 258 |
""")
|
| 259 |
|
| 260 |
+
with gr.TabItem("π€ Submission", elem_classes=["tab-item"]):
|
| 261 |
with gr.Row():
|
| 262 |
file_input = gr.File(label="π Upload Prediction CSV", file_types=[".csv"], interactive=True)
|
| 263 |
model_name_input = gr.Textbox(label="ποΈ Model Name", placeholder="Enter your model name")
|
|
|
|
| 275 |
outputs=[eval_status, overall_accuracy_display],
|
| 276 |
)
|
| 277 |
|
| 278 |
+
with gr.TabItem("π
Leaderboard", elem_classes=["tab-item"]):
|
| 279 |
leaderboard_table = gr.Dataframe(
|
| 280 |
value=load_leaderboard(),
|
| 281 |
label="Leaderboard",
|
|
|
|
| 292 |
gr.Markdown(f"**Last updated:** {LAST_UPDATED}")
|
| 293 |
|
| 294 |
demo.launch()
|
|
|
|
|
|