src/about.py

from dataclasses import dataclass
from enum import Enum

@dataclass(frozen=True)
class Task:
    benchmark: str
    metric: str
    col_name: str
    type: str
    baseline: float = 0.0


# 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 
    # task2 = Task("belebele_pol_Latn", "acc,none", "belebele_pol_Latn", "multiple_choice", 0.279)
    # task3 = Task("polemo2_in", "exact_match,score-first", "polemo2-in_g", "generate_until", 0.416)
    # task4 = Task("polemo2_in_multiple_choice", "acc,none", "polemo2-in_mc", "multiple_choice", 0.416)
    # task5 = Task("polemo2_out", "exact_match,score-first", "polemo2-out_g", "generate_until", 0.368)
    # task6 = Task("polemo2_out_multiple_choice", "acc,none", "polemo2-out_mc", "multiple_choice", 0.368)
    # task7 = Task("polish_8tags_multiple_choice", "acc,none", "8tags_mc", "multiple_choice", 0.143)
    # task8 = Task("polish_8tags_regex", "exact_match,score-first", "8tags_g", "generate_until", 0.143)
    # task9a = Task("polish_belebele_mc", "acc,none", "belebele_mc", "multiple_choice", 0.279)
    # task9 = Task("polish_belebele_regex", "exact_match,score-first", "belebele_g", "generate_until", 0.279)
    # task10 = Task("polish_dyk_multiple_choice", "f1,none", "dyk_mc", "multiple_choice", 0.289)
    # task11 = Task("polish_dyk_regex", "f1,score-first", "dyk_g", "generate_until", 0.289)
    # task12 = Task("polish_ppc_multiple_choice", "acc,none", "ppc_mc", "multiple_choice", 0.419)
    # task13 = Task("polish_ppc_regex", "exact_match,score-first", "ppc_g", "generate_until", 0.419)
    # task14 = Task("polish_psc_multiple_choice", "f1,none", "psc_mc", "multiple_choice", 0.466)
    # task15 = Task("polish_psc_regex", "f1,score-first", "psc_g", "generate_until", 0.466)
    # task16 = Task("polish_cbd_multiple_choice", "f1,none", "cbd_mc", "multiple_choice", 0.149)
    # task17 = Task("polish_cbd_regex", "f1,score-first", "cbd_g", "generate_until", 0.149)
    # task18 = Task("polish_klej_ner_multiple_choice", "acc,none", "klej_ner_mc", "multiple_choice", 0.343)
    # task19 = Task("polish_klej_ner_regex", "exact_match,score-first", "klej_ner_g", "generate_until", 0.343)
    # task21 = Task("polish_polqa_reranking_multiple_choice", "acc,none", "polqa_reranking_mc", "multiple_choice", 0.5335588952710677) # multiple_choice
    # task22 = Task("polish_polqa_open_book", "levenshtein,none", "polqa_open_book_g", "generate_until", 0.0) # generate_until
    # task23 = Task("polish_polqa_closed_book", "levenshtein,none", "polqa_closed_book_g", "generate_until", 0.0) # generate_until
    # task24 = Task("polish_poquad_open_book", "levenshtein,none", "poquad_open_book", "generate_until", 0.0)
    # task25 = Task("polish_eq_bench_first_turn", "first_eqbench,none", "eq_bench_first_turn", "generate_until", 0.0)
    # task26 = Task("polish_eq_bench", "average_eqbench,none", "eq_bench", "generate_until", 0.0)
    # task20 = Task("polish_poleval2018_task3_test_10k", "word_perplexity,none", "poleval2018_task3_test_10k", "other")
    # task27 = Task("polish_poquad_reranking", "acc,none", "poquad_reranking", "other", 0.0)
    # task28 = Task("polish_abstractive_poquad_rag", "levenshtein,none", "abstractive_poquad_rag", "other", 0.0)
    # task29 = Task("polish_abstractive_poquad_open_book", "levenshtein,none", "abstractive_poquad_open_book", "other", 0.0)
    task30a = Task("polish_pes", "exact_match,score-first", "pes", "generate_until", 0.0)
    task60 = Task("polish_pes_medycyna_rodzinna", "exact_match,score-first", "medycyna_rodzinna", "generate_until", 0.0)
    task73 = Task("polish_pes_pediatria", "exact_match,score-first", "pediatria", "generate_until", 0.0)

    task30 = Task("polish_pes_alergologia", "exact_match,score-first", "alergologia", "generate_until", 0.0)
    task31 = Task("polish_pes_anestezjologia", "exact_match,score-first", "anestezjologia", "generate_until", 0.0)
    task32 = Task("polish_pes_angiologia", "exact_match,score-first", "angiologia", "generate_until", 0.0)
    task33 = Task("polish_pes_balneologia_i_medycyna_fizykalna", "exact_match,score-first", "balneologia_i_medycyna_fizykalna", "generate_until", 0.0)
    task34 = Task("polish_pes_chirurgia_dziecieca", "exact_match,score-first", "chirurgia_dziecieca", "generate_until", 0.0)
    task35 = Task("polish_pes_chirurgia_naczyniowa", "exact_match,score-first", "chirurgia_naczyniowa", "generate_until", 0.0)
    task36 = Task("polish_pes_chirurgia_ogolna", "exact_match,score-first", "chirurgia_ogolna", "generate_until", 0.0)
    task37 = Task("polish_pes_chirurgia_onkologiczna", "exact_match,score-first", "chirurgia_onkologiczna", "generate_until", 0.0)
    task38 = Task("polish_pes_chirurgia_stomatologiczna", "exact_match,score-first", "chirurgia_stomatologiczna", "generate_until", 0.0)
    task39 = Task("polish_pes_chirurgia_szczekowo-twarzowa", "exact_match,score-first", "chirurgia_szczekowo-twarzowa", "generate_until", 0.0)
    task40 = Task("polish_pes_choroby_pluc", "exact_match,score-first", "choroby_pluc", "generate_until", 0.0)
    task41 = Task("polish_pes_choroby_pluc_dzieci", "exact_match,score-first", "choroby_pluc_dzieci", "generate_until", 0.0)
    task42 = Task("polish_pes_choroby_wewnetrzne", "exact_match,score-first", "choroby_wewnetrzne", "generate_until", 0.0)
    task43 = Task("polish_pes_choroby_zakazne", "exact_match,score-first", "choroby_zakazne", "generate_until", 0.0)
    task44 = Task("polish_pes_dermatologia_i_wenerologia", "exact_match,score-first", "dermatologia_i_wenerologia", "generate_until", 0.0)
    task45 = Task("polish_pes_diabetologia", "exact_match,score-first", "diabetologia", "generate_until", 0.0)
    task46 = Task("polish_pes_endokrynologia", "exact_match,score-first", "endokrynologia", "generate_until", 0.0)
    task47 = Task("polish_pes_endokrynologia_ginekologiczna_i_rozrodczosc", "exact_match,score-first", "endokrynologia_ginekologiczna_i_rozrodczosc", "generate_until", 0.0)
    task48 = Task("polish_pes_endokrynologia_i_diabetologia_dziecieca", "exact_match,score-first", "endokrynologia_i_diabetologia_dziecieca", "generate_until", 0.0)
    task49 = Task("polish_pes_gastroenterologia", "exact_match,score-first", "gastroenterologia", "generate_until", 0.0)
    task50 = Task("polish_pes_gastroenterologia_dziecieca", "exact_match,score-first", "gastroenterologia_dziecieca", "generate_until", 0.0)
    task51 = Task("polish_pes_geriatria", "exact_match,score-first", "geriatria", "generate_until", 0.0)
    task52 = Task("polish_pes_ginekologia_onkologiczna", "exact_match,score-first", "ginekologia_onkologiczna", "generate_until", 0.0)
    task53 = Task("polish_pes_hematologia", "exact_match,score-first", "hematologia", "generate_until", 0.0)
    task54 = Task("polish_pes_hipertensjologia", "exact_match,score-first", "hipertensjologia", "generate_until", 0.0)
    task55 = Task("polish_pes_kardiochirurgia", "exact_match,score-first", "kardiochirurgia", "generate_until", 0.0)
    task56 = Task("polish_pes_kardiologia", "exact_match,score-first", "kardiologia", "generate_until", 0.0)
    task57 = Task("polish_pes_medycyna_pracy", "exact_match,score-first", "medycyna_pracy", "generate_until", 0.0)
    task58 = Task("polish_pes_medycyna_paliatywna", "exact_match,score-first", "medycyna_paliatywna", "generate_until", 0.0)
    task59 = Task("polish_pes_medycyna_ratunkowa", "exact_match,score-first", "medycyna_ratunkowa", "generate_until", 0.0)

    task61 = Task("polish_pes_medycyna_sportowa", "exact_match,score-first", "medycyna_sportowa", "generate_until", 0.0)
    task62 = Task("polish_pes_nefrologia", "exact_match,score-first", "nefrologia", "generate_until", 0.0)
    task63 = Task("polish_pes_neonatologia", "exact_match,score-first", "neonatologia", "generate_until", 0.0)
    task64 = Task("polish_pes_neurochirurgia", "exact_match,score-first", "neurochirurgia", "generate_until", 0.0)
    task65 = Task("polish_pes_neurologia", "exact_match,score-first", "neurologia", "generate_until", 0.0)
    task66 = Task("polish_pes_neurologia_dziecieca", "exact_match,score-first", "neurologia_dziecieca", "generate_until", 0.0)
    task67 = Task("polish_pes_okulistyka", "exact_match,score-first", "okulistyka", "generate_until", 0.0)
    task68 = Task("polish_pes_onkologia_kliniczna", "exact_match,score-first", "onkologia_kliniczna", "generate_until", 0.0)
    task69 = Task("polish_pes_ortodoncja", "exact_match,score-first", "ortodoncja", "generate_until", 0.0)
    task70 = Task("polish_pes_ortopedia", "exact_match,score-first", "ortopedia", "generate_until", 0.0)
    task71 = Task("polish_pes_otolaryngologia", "exact_match,score-first", "otolaryngologia", "generate_until", 0.0)
    task72 = Task("polish_pes_patomorfologia", "exact_match,score-first", "patomorfologia", "generate_until", 0.0)
    
    task74 = Task("polish_pes_perinatologia", "exact_match,score-first", "perinatologia", "generate_until", 0.0)
    task75 = Task("polish_pes_periodontologia", "exact_match,score-first", "periodontologia", "generate_until", 0.0)
    task76 = Task("polish_pes_poloznictwo_i_ginekologia", "exact_match,score-first", "poloznictwo_i_ginekologia", "generate_until", 0.0)
    task77 = Task("polish_pes_protetyka_stomatologiczna", "exact_match,score-first", "protetyka_stomatologiczna", "generate_until", 0.0)
    task78 = Task("polish_pes_psychiatria", "exact_match,score-first", "psychiatria", "generate_until", 0.0)
    task79 = Task("polish_pes_psychiatria_dzieci_i_mlodziezy", "exact_match,score-first", "psychiatria_dzieci_i_mlodziezy", "generate_until", 0.0)
    task80 = Task("polish_pes_radiologia_i_diagnostyka_obrazowa", "exact_match,score-first", "radiologia_i_diagnostyka_obrazowa", "generate_until", 0.0)
    task81 = Task("polish_pes_radioterapia_onkologiczna", "exact_match,score-first", "radioterapia_onkologiczna", "generate_until", 0.0)
    task82 = Task("polish_pes_rehabilitacja_medyczna", "exact_match,score-first", "rehabilitacja_medyczna", "generate_until", 0.0)
    task83 = Task("polish_pes_reumatologia", "exact_match,score-first", "reumatologia", "generate_until", 0.0)
    task84 = Task("polish_pes_stomatologia_dziecieca", "exact_match,score-first", "stomatologia_dziecieca", "generate_until", 0.0)
    task85 = Task("polish_pes_stomatologia_zachowawcza", "exact_match,score-first", "stomatologia_zachowawcza", "generate_until", 0.0)
    task86 = Task("polish_pes_transplantologia_kliniczna", "exact_match,score-first", "transplantologia_kliniczna", "generate_until", 0.0)


g_tasks = [task.value.benchmark for task in Tasks if task.value.type == "generate_until"]
mc_tasks = [task.value.benchmark for task in Tasks if task.value.type == "multiple_choice"]
rag_tasks = ['polish_polqa_reranking_multiple_choice', 'polish_polqa_open_book', 'polish_poquad_open_book']
all_tasks = g_tasks + mc_tasks
all_tasks.remove('polish_pes')

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



# Your leaderboard name
TITLE = """
<div style="display: flex; flex-wrap: wrap; justify-content: space-around;">
    <img src="https://speakleash.org/wp-content/uploads/2023/09/SpeakLeash_logo.svg">
    <div>
        <h1 align="center" id="space-title">Polish Medical Leaderboard</h1>
        <h2 align="center" id="space-subtitle">Leaderboard was created as part of an open-science project SpeakLeash.org</h2>
    </div>
</div>
"""

# What does your leaderboard evaluate?
INTRODUCTION_TEXT = f"""
The leaderboard evaluates language models on Polish Board Certification Examinations (Państwowy Egzamin Specjalizacyjny) from years 2018-2022.

Average columns are normalized against scores by "Baseline (majority class)".

* `,chat` suffix means that a model is tested using chat templates
* `,chat,multiturn` suffix means that a model is tested using chat templates and fewshot examples are treated as a multi-turn conversation
* 🚧 prefix means that not all tasks were calculated and the average scores are not accurate

We gratefully acknowledge Polish high-performance computing infrastructure PLGrid (HPC Centers: ACK Cyfronet AGH) for providing computer facilities and support within computational grant no. PLG/2024/016951.
"""

# Which evaluations are you running? how can people reproduce what you have?
LLM_BENCHMARKS_TEXT = f"""
The benchmark uses datasets:
* [speakleash/PES-2018-2022](https://huggingface.co/datasets/speakleash/PES-2018-2022), which is based on [amu-cai/PES-2018-2022](https://huggingface.co/datasets/amu-cai/PES-2018-2022) ([J. Pokrywka, J. Kaczmarek, E. Gorzelańczyk, GPT-4 passes most of the 297 written Polish Board Certification Examinations, 2024](https://arxiv.org/abs/2405.01589)).

## Do you want to add your model to the leaderboard?

Contact with me: [LinkedIn](https://www.linkedin.com/in/wrobelkrzysztof/)

or join our [Discord SpeakLeash](https://discord.gg/FfYp4V6y3R)

## TODO

* fix long model names
* add inference time
* add more tasks
* fix scrolling on Firefox

## Tasks

Tasks taken into account while calculating averages:
* Average: {', '.join(all_tasks)}


## Reproducibility
To reproduce our results, you need to clone the repository:

```
git clone https://github.com/speakleash/lm-evaluation-harness.git -b polish4
cd lm-evaluation-harness
pip install -e .
```

and run benchmark for 0-shot and 5-shot:

```
lm_eval --model hf --model_args pretrained=speakleash/Bielik-7B-Instruct-v0.1 --tasks polish_pes --num_fewshot 0 --output_path results/ --log_samples
lm_eval --model hf --model_args pretrained=speakleash/Bielik-7B-Instruct-v0.1 --tasks polish_pes --num_fewshot 5 --output_path results/ --log_samples
```

With chat templates:
```
lm_eval --model hf --model_args pretrained=speakleash/Bielik-7B-Instruct-v0.1 --tasks polish_pes --num_fewshot 0 --output_path results/ --log_samples --apply_chat_template
lm_eval --model hf --model_args pretrained=speakleash/Bielik-7B-Instruct-v0.1 --tasks polish_pes --num_fewshot 5 --output_path results/ --log_samples --apply_chat_template
```

## List of Polish models

* speakleash/Bielik-11B-v2.2-Instruct
* speakleash/Bielik-11B-v2.1-Instruct
* speakleash/Bielik-11B-v2.0-Instruct
* speakleash/Bielik-7B-Instruct-v0.1

### List of multilingual models

* Meta-Llama-3.1-405B-Instruct-FP8
* Mistral-Large-Instruct-2407
* Meta-Llama-3.1-70B-Instruct
* Qwen2-72B-Instruct
* Meta-Llama-3-70B-Instruct
* glm-4-9b-chat
* Meta-Llama-3.1-8B-Instruct
* Mistral-Nemo-Instruct-2407-PL-finetuned
* Mistral-Nemo-Instruct-2407
* Qwen2-7B-Instruct
* Meditron3-8B
* SOLAR-10.7B-Instruct-v1.0
* openchat-3.5-0106-gemma
* Starling-LM-7B-beta
* Mistral-7B-Instruct-v0.3
* Phi-3.5-mini-instruct

"""

EVALUATION_QUEUE_TEXT = """
## Some good practices before submitting a model

### 1) Make sure you can load your model and tokenizer using AutoClasses:
```python
from transformers import AutoConfig, AutoModel, AutoTokenizer
config = AutoConfig.from_pretrained("your model name", revision=revision)
model = AutoModel.from_pretrained("your model name", revision=revision)
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)
```
If this step fails, follow the error messages to debug your model before submitting it. It's likely your model has been improperly uploaded.

Note: make sure your model is public!
Note: if your model needs `use_remote_code=True`, we do not support this option yet but we are working on adding it, stay posted!

### 2) Convert your model weights to [safetensors](https://huggingface.co/docs/safetensors/index)
It's a new format for storing weights which is safer and faster to load and use. It will also allow us to add the number of parameters of your model to the `Extended Viewer`!

### 3) Make sure your model has an open license!
This is a leaderboard for Open LLMs, and we'd love for as many people as possible to know they can use your model 🤗

### 4) Fill up your model card
When we add extra information about models to the leaderboard, it will be automatically taken from the model card

## In case of model failure
If your model is displayed in the `FAILED` category, its execution stopped.
Make sure you have followed the above steps first.
If everything is done, check you can launch the EleutherAIHarness on your model locally, using the above command without modifications (you can add `--limit` to limit the number of examples per task).
"""

CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
CITATION_BUTTON_TEXT = r"""@misc{polish-medical-llm-leaderboard,
	title        = {Polish Medical Leaderboard},
	author       = {Wróbel, Krzysztof and {SpeakLeash Team} and {Cyfronet Team}},
	year         = 2024,
	publisher    = {Hugging Face},
	howpublished = "\url{https://huggingface.co/spaces/speakleash/polish_medical_leaderboard}"
}

@misc{eval-harness,
  author       = {Gao, Leo and Tow, Jonathan and Abbasi, Baber and Biderman, Stella and Black, Sid and DiPofi, Anthony and Foster, Charles and Golding, Laurence and Hsu, Jeffrey and Le Noac'h, Alain and Li, Haonan and McDonell, Kyle and Muennighoff, Niklas and Ociepa, Chris and Phang, Jason and Reynolds, Laria and Schoelkopf, Hailey and Skowron, Aviya and Sutawika, Lintang and Tang, Eric and Thite, Anish and Wang, Ben and Wang, Kevin and Zou, Andy},
  title        = {A framework for few-shot language model evaluation},
  month        = 07,
  year         = 2024,
  publisher    = {Zenodo},
  version      = {v0.4.3},
  doi          = {10.5281/zenodo.12608602},
  url          = {https://zenodo.org/records/12608602}
}

@misc{fwpes2024,
  title={speakleash/PES-2018-2022},
  author={Maria, Filipkowska and Krzyszof, Wróbel},
  year={2024},
  howpublished={\url{https://huggingface.co/datasets/speakleash/PES-2018-2022}},
}

@misc{pkgpes2024,
  title={amu-cai/PES-2018-2022},
  author={Jakub Pokrywka and Jeremi Kaczmarek and Edward Gorzelańczyk},
  year={2024},
  howpublished={\url{https://huggingface.co/datasets/amu-cai/PES-2018-2022}},
}

@misc{pokrywka2024gpt4,
      title={GPT-4 passes most of the 297 written Polish Board Certification Examinations}, 
      author={Jakub Pokrywka and Jeremi Kaczmarek and Edward Gorzelańczyk},
      year={2024},
      eprint={2405.01589},
      archivePrefix={arXiv},
      primaryClass={id='cs.CL' full_name='Computation and Language' is_active=True alt_name='cmp-lg' in_archive='cs' is_general=False description='Covers natural language processing. Roughly includes material in ACM Subject Class I.2.7. Note that work on artificial languages (programming languages, logics, formal systems) that does not explicitly address natural-language issues broadly construed (natural-language processing, computational linguistics, speech, text retrieval, etc.) is not appropriate for this area.'}
}

@misc{open-pl-llm-leaderboard,
	title        = {Open PL LLM Leaderboard},
	author       = {Wróbel, Krzysztof and {SpeakLeash Team} and {Cyfronet Team}},
	year         = 2024,
	publisher    = {Hugging Face},
	howpublished = "\url{https://huggingface.co/spaces/speakleash/open_pl_llm_leaderboard}"
}

@misc{open-llm-leaderboard-v1,
  author = {Edward Beeching and Clémentine Fourrier and Nathan Habib and Sheon Han and Nathan Lambert and Nazneen Rajani and Omar Sanseviero and Lewis Tunstall and Thomas Wolf},
  title = {Open LLM Leaderboard (2023-2024)},
  year = {2023},
  publisher = {Hugging Face},
  howpublished = "\url{https://huggingface.co/spaces/open-llm-leaderboard-old/open_llm_leaderboard}"
}
"""