src/about.py

from dataclasses import dataclass
from enum import Enum


@dataclass
class Task:
    benchmark: str
    metric: str
    col_name: str
    category: str
    higher_is_better: bool = True
    scale_by_100: bool = True


# Select your tasks here
# ---------------------------------------------------
class Tasks(Enum):
    # task_key in the json file, metric_key in the json file, name to display in the leaderboard
    task1 = Task("ami_2020_aggressiveness", "f1,none", "AMI 2020 Agg", "NLU")
    task2 = Task("ami_2020_misogyny", "f1,none", "AMI 2020 Miso", "NLU")
    task0 = Task("arc_challenge_ita", "acc_norm,none", "ARC-C", "CFK")
    task4 = Task("belebele_ita", "acc_norm,none", "Belebele", "NLU")
    task3 = Task("gente_rephrasing", "acc,none", "GeNTE Neutralizing", "BFS")
    task12 = Task("haspeede2_hs", "f1,none", "HaSpeeDe2 HS", "BFS")
    task13 = Task("haspeede2_stereo", "f1,none", "HaSpeeDe2 Stereo", "BFS")
    task5 = Task("hatecheck_ita", "f1,none", "HateCheck", "BFS")
    task6 = Task("honest_ita", "acc,none", "HONEST", "BFS", higher_is_better=False)
    task14 = Task("ironita_irony", "f1,none", "IronITA Irony", "NLU")
    task15 = Task("ironita_sarcasm", "f1,none", "IronITA Sarcasm", "NLU")
    task7 = Task("itacola", "mcc,none", "ItaCoLA", "NLU", scale_by_100=False)
    task8 = Task("news_sum", "bertscore,none", "News Sum", "NLU")
    task16 = Task("sentipolc", "f1,none", "SENTIPOLC", "NLU")
    task9 = Task("squad_it", "squad_f1,get-answer", "SQuAD it", "CFK")
    task10 = Task("truthfulqa_mc2_ita", "acc,none", "TruthfulQA", "CFK")
    task11 = Task("xcopa_it", "acc,none", "XCOPA", "CFK")
    task17 = Task("hellaswag_ita", "acc_norm,none", "Hellaswag-it", "CFK")


NUM_FEWSHOT = 0  # Change with your few shot
# ---------------------------------------------------


# Your leaderboard name
TITLE = """<h1 align="center" id="space-title">ItaEval leaderboard</h1>"""

# What does your leaderboard evaluate?
INTRODUCTION_TEXT = """
This leaderboard evaluates language models on <b>ItaEval</b>, a new unified benchmark for Italian.

Submit your model: [Google Form](https://forms.gle/xpGH66DpVRcCmdcJ6)

Some information:
- Unlike other leaderboards you may find online, we do not support automatic evaluation for new model submissions. Currently, we are taking care of running models on the suite. Please fill out the form above to have your model evaluated and included here.
- You can find some more details on the suite in our [technical report](https://bit.ly/itaeval_tweetyita_v1)
"""

ITA_EVAL_REPO = "https://github.com/g8a9/ita-eval"

# Which evaluations are you running? how can people reproduce what you have?
LLM_BENCHMARKS_TEXT = f"""
## How it works

## Reproducibility
Head to {ITA_EVAL_REPO} to reproduce our results for all the instructions.

If all the setup goes smoothly, you can run 'MODEL' on ItaEval with:
```bash
MODEL="your-model-id-on-the-huggingface-hub"
lm_eval --model hf \
    --model_args pretrained=$MODEL,dtype=bfloat16 \
    --tasks ita_eval \
    --batch_size 1 \
    --log_samples \
    --output_path "."
```
"""

EVALUATION_QUEUE_TEXT = """
We do not plan to accept autonomous submissions, yet. Fill [this form](https://forms.gle/xpGH66DpVRcCmdcJ6) to have your model evaluated. 
"""

CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
CITATION_BUTTON_TEXT = r"""
We are working on it! :)
"""