Datasetpublic

MINGYISU/mscoco_omni

catalog.jsonl contains the captions and filenames for each image id. There are around 170 text-image-video-audio (omni) tuples in it. How to run Specify GOOGLE_API_KEY run the following to get setup readybash setup.sh then only need to run this file only in the futurepython geminiAPI.py If runs successfully, a file called mscoco_cmret.jsonl will be generated, please provide this file to me.

sourceHugging Facecc-by-4.0updated 4mo agoView on Hugging Face

0likes81downloads

Dataset Card

# catalog.jsonl contains the captions and filenames for each image id. There are around 170 text-image-video-audio (omni) tuples in it.

How to run

1.Specify GOOGLEAPIKEY
2.run the following to get setup ready

bash

   bash setup.sh

1.then only need to run this file only in the future

bash

   python geminiAPI.py

1.If runs successfully, a file called mscoco_cmret.jsonl will be generated, please provide this file to me.