update utils
Browse files- src/about.py +10 -5
src/about.py
CHANGED
@@ -8,6 +8,7 @@ class TaskInfo:
|
|
8 |
col_name: str
|
9 |
metric: str
|
10 |
|
|
|
11 |
# Replace these with actual subjects from your dataset
|
12 |
class Tasks(Enum):
|
13 |
History = TaskInfo(benchmark='History', col_name='History', metric='accuracy')
|
@@ -20,15 +21,15 @@ class Tasks(Enum):
|
|
20 |
Chemistry = TaskInfo(benchmark='Chemistry', col_name='Chemistry', metric='accuracy')
|
21 |
Biology = TaskInfo(benchmark='Biology', col_name='Biology', metric='accuracy')
|
22 |
ComputerScience = TaskInfo(benchmark='Computer Science', col_name='Computer Science', metric='accuracy')
|
23 |
-
|
24 |
# Now include the variables expected by app.py
|
25 |
|
26 |
TITLE = """
|
27 |
-
<h1 align="center">π
|
28 |
"""
|
29 |
|
30 |
INTRODUCTION_TEXT = """
|
31 |
-
Welcome to the
|
32 |
"""
|
33 |
|
34 |
LLM_BENCHMARKS_TEXT = """
|
@@ -54,6 +55,10 @@ Below are the lists of models that have been evaluated, are currently being eval
|
|
54 |
CITATION_BUTTON_LABEL = "Citation"
|
55 |
CITATION_BUTTON_TEXT = """
|
56 |
If you use this leaderboard or the MMMLU dataset in your research, please cite:
|
57 |
-
@
|
58 |
-
|
|
|
|
|
|
|
|
|
59 |
|
|
|
8 |
col_name: str
|
9 |
metric: str
|
10 |
|
11 |
+
'''
|
12 |
# Replace these with actual subjects from your dataset
|
13 |
class Tasks(Enum):
|
14 |
History = TaskInfo(benchmark='History', col_name='History', metric='accuracy')
|
|
|
21 |
Chemistry = TaskInfo(benchmark='Chemistry', col_name='Chemistry', metric='accuracy')
|
22 |
Biology = TaskInfo(benchmark='Biology', col_name='Biology', metric='accuracy')
|
23 |
ComputerScience = TaskInfo(benchmark='Computer Science', col_name='Computer Science', metric='accuracy')
|
24 |
+
'''
|
25 |
# Now include the variables expected by app.py
|
26 |
|
27 |
TITLE = """
|
28 |
+
<h1 align="center">π Arabic MMMLU Evaluation Leaderboard for LLMs π</h1>
|
29 |
"""
|
30 |
|
31 |
INTRODUCTION_TEXT = """
|
32 |
+
Welcome to the Arabic MMMLU Evaluation for LLMs Leaderboard for the MMMLU dataset evaluation. This leaderboard displays the performance of various language models on the MMMLU dataset across different subjects.
|
33 |
"""
|
34 |
|
35 |
LLM_BENCHMARKS_TEXT = """
|
|
|
55 |
CITATION_BUTTON_LABEL = "Citation"
|
56 |
CITATION_BUTTON_TEXT = """
|
57 |
If you use this leaderboard or the MMMLU dataset in your research, please cite:
|
58 |
+
@misc{AMMMLU,
|
59 |
+
author = {Nacar, Omer},
|
60 |
+
title = {Arabic MMMLU Evaluation for LLMs Leaderboard},
|
61 |
+
year = {2024},
|
62 |
+
publisher = {Omartificial-Intelligence-Space}}"
|
63 |
+
}"""
|
64 |
|