Clémentine
POC, v0
84b5dfa
raw
history blame
1.79 kB
from src.static.tag_info import *
TITLE = "# Leaderboard explorer"
INTRO = """
Have you ever wondered which leaderboard would be best for your use case?
"""
ABOUT = ("""
If you want your leaderboard to appear in our suggestions, feel free to add relevant information in its tag metadata, and it will be displayed here.
# First step
Make sure to either use the tag `leaderboard` or `arena` to your space, by adding the following to your README
```
tags:
- leaderboard
```
# Extra tags
## Submission type
Arenas are not concerned by this category.
""" +
"\n".join([f"- {s.value.key}: {s.value.usage}" for s in SubmissionType]) +
"""
## Test set status
Arenas are not concerned by this category.
""" +
"\n".join([f"- {s.value.key}: {s.value.usage}" for s in TestSetStatus]) +
"""
## Judges
""" +
"\n".join([f"- {s.value.key}: {s.value.usage}" for s in Judge]) +
"""
## Modalities
Can be any (or several) of the following list:
""" +
"\n".join([f"- {s.value.key}: {s.value.usage}" for s in Modality]) +
"""
## Evaluation categories
Can be any (or several) of the following list:
""" +
"\n".join([f"- {s.value.key}: {s.value.usage}" for s in EvaluationCategory]) +
"""
## Language
You can indicate the languages covered by your benchmark like so: `language:mylanguage`.
At the moment, we do not support language codes, please use the language name in English.
""")
DOCUMENTATION = """
How to create your own leaderboard?
I'll make an updated documentation page here at some point, but for now, you can check our [demo leaderboard org](https://huggingface.co/demo-leaderboard-backend)!
You just need to duplicate the front space (and backend if you want to run your leaderboard on spaces compute), copy the datasets to your own org, and edit the env variables.
"""