Doubt regarding dataset[DOUBT] <"Your doubt goes here">

**Describe the doubt**
A clear and concise description of what the doubt is.
Dataset is very large and downloading it from GitHub takes forever. After that again uploading it on Kaggle or g-drive(for colab)
because at the end of the day training needs GPU, doing this is too much difficult because if it gets interrupted than start again.

**Where do you need help**
Where exactly do you need our help?
If possible upload it on Kaggle and provide the link so that it makes it easy to use the dataset.

**Additional context**
Add any other context about the doubt here.
If you considering this than try to upload with missing images.

Start by locating the large dataset currently referenced by the project and checking how it is downloaded from GitHub. Compare the complete dataset with the suggested missing-images version, then verify that a Kaggle or Google Drive link supports the project's GPU/Colab training workflow without requiring a restart after interruption.

Written by the indexing model from the issue text.