| ---
|
| license: cc-by-4.0
|
| tags:
|
| - adaptive
|
| - architecture
|
| - jsonl
|
| - leave-one-out
|
| - mixup-cutmix
|
| - pointcloud-text
|
| - self-training
|
| - standard
|
| - stratified
|
| ---
|
|
|
| # dataset_055326048_architecture_pointcloud_text.py
|
|
|
| ## Dataset Summary
|
|
|
| A **architecture** dataset with **pointcloud text** modality, stored in **jsonl** format.
|
|
|
| ## Preprocessing & Augmentation
|
|
|
| - **Preprocessing**: standard
|
| - **Augmentation**: mixup cutmix
|
|
|
| ## Splits & Sampling
|
|
|
| - **Split strategy**: leave one out
|
| - **Sampling**: stratified
|
|
|
| ## Quality & Labeling
|
|
|
| - **Quality filtering**: adaptive
|
| - **Labeling**: self training
|
|
|
| ## Files
|
|
|
| - `dataset_055326048_architecture_pointcloud_text.py` — main artifact of this repository
|
|
|
| ## License
|
|
|
| See the license field above.
|
|
|
