OpenCompass/docs/en/get_started/installation.md
Xingjun.Wang edab1c07ba
[Feature] Support ModelScope datasets (#1289)
* add ceval, gsm8k modelscope surpport

* update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest

* update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets

* format file

* format file

* update dataset format

* support ms_dataset

* udpate dataset for modelscope support

* merge myl_dev and update test_ms_dataset

* udpate dataset for modelscope support

* update readme

* update eval_api_zhipu_v2

* remove unused code

* add get_data_path function

* update readme

* remove tydiqa japanese subset

* add ceval, gsm8k modelscope surpport

* update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest

* update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets

* format file

* format file

* update dataset format

* support ms_dataset

* udpate dataset for modelscope support

* merge myl_dev and update test_ms_dataset

* update readme

* udpate dataset for modelscope support

* update eval_api_zhipu_v2

* remove unused code

* add get_data_path function

* remove tydiqa japanese subset

* update util

* remove .DS_Store

* fix md format

* move util into package

* update docs/get_started.md

* restore eval_api_zhipu_v2.py, add environment setting

* Update dataset

* Update

* Update

* Update

* Update

---------

Co-authored-by: Yun lin <yunlin@U-Q9X2K4QV-1904.local>
Co-authored-by: Yunnglin <mao.looper@qq.com>
Co-authored-by: Yun lin <yunlin@laptop.local>
Co-authored-by: Yunnglin <maoyl@smail.nju.edu.cn>
Co-authored-by: zhangsongyang <zhangsongyang@pjlab.org.cn>
2024-07-29 13:48:32 +08:00

4.7 KiB
Raw Blame History

Installation

  1. Set up the OpenCompass environment:
````{tab} Open-source Models with GPU

   ```bash
   conda create --name opencompass python=3.10 pytorch torchvision pytorch-cuda -c nvidia -c pytorch -y
   conda activate opencompass
   ```

   If you want to customize the PyTorch version or related CUDA version, please refer to the [official documentation](https://pytorch.org/get-started/locally/) to set up the PyTorch environment. Note that OpenCompass requires `pytorch>=1.13`.

````
````{tab} API Models with CPU-only

   ```bash
   conda create -n opencompass python=3.10 pytorch torchvision torchaudio cpuonly -c pytorch -y
   conda activate opencompass
   # also please install requiresments packages via `pip install -r requirements/api.txt` for API models if needed.
   ```

   If you want to customize the PyTorch version, please refer to the [official documentation](https://pytorch.org/get-started/locally/) to set up the PyTorch environment. Note that OpenCompass requires `pytorch>=1.13`.

````
  1. Install OpenCompass:

    git clone https://github.com/open-compass/opencompass.git
    cd opencompass
    pip install -e .
    
  2. Install humaneval (Optional)

    If you want to evaluate your models coding ability on the humaneval dataset, follow this step.

    click to show the details
    git clone https://github.com/openai/human-eval.git
    cd human-eval
    pip install -r requirements.txt
    pip install -e .
    cd ..
    

    Please read the comments in human_eval/execution.py lines 48-57 to understand the potential risks of executing the model generation code. If you accept these risks, uncomment line 58 to enable code execution evaluation.

  3. Install Llama (Optional)

    If you want to evaluate Llama / Llama-2 / Llama-2-chat with its official implementation, follow this step.

    click to show the details
    git clone https://github.com/facebookresearch/llama.git
    cd llama
    pip install -r requirements.txt
    pip install -e .
    cd ..
    

    You can find example configs in configs/models. (example)

  4. Install alpaca-eval (Optional)

    If you want toevaluate alpaca-eval in official ways, follow this step.

    click to show the details
    pip install alpaca-eval
    

Dataset Preparation

The datasets supported by OpenCompass mainly include three parts:

  1. Huggingface datasets: The Huggingface Datasets provide a large number of datasets, which will automatically download when running with this option. Translate the paragraph into English:

  2. ModelScope Datasets: ModelScope OpenCompass Dataset supports automatic downloading of datasets from ModelScope.

    To enable this feature, set the environment variable: export DATASET_SOURCE=ModelScope. The available datasets include (sourced from OpenCompassData-core.zip):

    humaneval, triviaqa, commonsenseqa, tydiqa, strategyqa, cmmlu, lambada, piqa, ceval, math, LCSTS, Xsum, winogrande, openbookqa, AGIEval, gsm8k, nq, race, siqa, mbpp, mmlu, hellaswag, ARC, BBH, xstory_cloze, summedits, GAOKAO-BENCH, OCNLI, cmnli
    
  3. Custom dataset: OpenCompass also provides some Chinese custom self-built datasets. Please run the following command to manually download and extract them.

Run the following commands to download and place the datasets in the ${OpenCompass}/data directory can complete dataset preparation.

# Run in the OpenCompass directory
wget https://github.com/open-compass/opencompass/releases/download/0.2.2.rc1/OpenCompassData-core-20240207.zip
unzip OpenCompassData-core-20240207.zip

If you need to use the more comprehensive dataset (~500M) provided by OpenCompass, You can download and unzip it using the following command:

wget https://github.com/open-compass/opencompass/releases/download/0.2.2.rc1/OpenCompassData-complete-20240207.zip
unzip OpenCompassData-complete-20240207.zip
cd ./data
find . -name "*.zip" -exec unzip "{}" \;

The list of datasets included in both .zip can be found here

OpenCompass has supported most of the datasets commonly used for performance comparison, please refer to configs/dataset for the specific list of supported datasets.

For next step, please read Quick Start.