Skip to content

OpenDCAI/DataFlow-MM

Repository files navigation

Dataflow-MM

Documents Ask DeepWiki

🎉 If you like our project, please give us a star ⭐ on GitHub for the latest update.

简体中文 | English

Quick Start

Install with the following command:

cd ./Dataflow-MM
conda create -n Dataflow-MM python=3.12
pip install -e .

Audio Test

Extra environments:

pip install -e ".[audio]"
pip install -e ".[vllm]"

测试命令

python /data0/gty/DataFlow-MM/test/test_whisper_promptedvqa.py
python /data0/gty/DataFlow-MM/test/test_audio_promptedvqa.py

python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_merge.py
python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_ctc_forced_aligner_filter.py
python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_ctc_forced_aligner.py
python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_silero_vad_generator.py
python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_whisper_promptedaqa.py
python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_promptedaqa.py
python /mnt/public/data/guotianyu/dataflow_project/DataFlow-MM/test/test_audio_asr_pipeline.py

nano-banana (gemini-v2.5-image) Test

测试命令

python test/test_image_editing.py --api_key < your api key >

we utilize the api from yucha

多参考图生成测试

测试命令

python test/test_echo4o_w_nano.py --api_key < your api key >

we utilize the api from yucha

About

Dataflow-MM, multi-media operators for Dataflow. We aim to prepare data for multimedia cases.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 9

Languages