from dataclasses import dataclass from enum import Enum @dataclass class Task: benchmark: str metric: str col_name: str # 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 # task0 = Task("anli_r1", "acc", "ANLI") # task1 = Task("logiqa", "acc_norm", "LogiQA") # retrieval tasks mrr = Task("retrieval", "mrr", "MRR ⬆️") map = Task("retrieval", "map", "MAP ⬆️") # generation tasks em = Task("generation", "em", "EM ⬆️") f1 = Task("generation", "f1", "F1 ⬆️") rouge1 = Task("generation", "rouge1", "Rouge-1 ⬆️") rouge2 = Task("generation", "rouge2", "Rouge-2 ⬆️") rougeL = Task("generation", "rougeL", "Rouge-L ⬆️") accuracy = Task("generation", "accuracy", "ACC ⬆️") completeness = Task("generation", "completeness", "COMP ⬆️") hallucination = Task("generation", "hallucination", "HAL ⬇️") utilization = Task("generation", "utilization", "UTIL ⬆️") numerical_accuracy = Task("generation", "numerical_accuracy", "MACC ⬆️") NUM_FEWSHOT = 0 # Change with your few shot # --------------------------------------------------- # Your leaderboard name TITLE = """

🏅 OmniEval Leaderboard

""" # What does your leaderboard evaluate? INTRODUCTION_TEXT = """

OmniEval: Omnidirectional and Automatic RAG Evaluation Benchmark in Financial Domain

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

OmniEval: Omnidirectional and Automatic RAG Evaluation Benchmark in Financial Domain

Installation | Quick-Start | License | Citation

## 🔧 Installation `conda env create -f environment.yml && conda activate finrag` ## 🚀 Quick-Start Notion: 1. The code run path is `./OpenFinBench` 2. We provide our auto-generated evaluation dataset in

### 1. Build the Retrieval Corpus ``` # cd OpenFinBench sh corpus_builder/build_corpus.sh # Please see the annotation inner the bash file to set parameters. ``` ### 2. Generate Evaluation Data Samples 1. Generate evaluation instances ``` # cd OpenFinBench sh data_generator/generate_data.sh ``` 2. Filter (quality inspection) evaluation instances ``` sh data_generator/generate_data_filter.sh ``` ### 3. Inference Your Models ``` # cd OpenFinBench sh evaluator/inference/rag_inference.sh ``` ### 4. Evaluate Your Models #### (a) Rule-based Evaluation ``` # cd OpenFinBench sh evaluator/judgement/judger.sh # by setting judge_type="rule" ``` #### (b) Model-based Evalution We propose five model-based metric: accuracy, completeness, utilization, numerical_accuracy, and hallucination. We have trained two models from Qwen2.5-7B by the lora strategy and human-annotation labels to implement model-based evaluation. Note that the evaluator of hallucination is different from other four. Their model checkpoint can be load from the following huggingface links: 1. The evaluator for hallucination metric:

2. The evaluator for other metric:

To implement model-based evaluation, you can first set up two vllm servers by the following codes: ``` ``` Then conduct the model-based evaluate using the following codes, (change the parameters inner the bash file). ``` sh evaluator/judgement/judger.sh ``` ## 🔖 License OmniEval is licensed under the [MIT License](./LICENSE). ## 🌟 Citation The paper is waiting to be released! """ 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""" """