|
Download README.md from HawkEyesAI/UBL-DisplayAudit: direct link, hf CLI and curl.
- Browser
- Download file 1.09 kB
-
https://huggingface.co/datasets/HawkEyesAI/UBL-DisplayAudit/resolve/main/README.md
- Command line
-
hf download hf://datasets/HawkEyesAI/UBL-DisplayAudit/README.md
-
curl -L -H "Authorization: Bearer $HF_TOKEN" -o README.md https://huggingface.co/datasets/HawkEyesAI/UBL-DisplayAudit/resolve/main/README.md
1.09 kB
# UBL Display Audit
Display audit dataset for UBL retail visibility and POSM analysis.
At a Glance
| Metric | Value |
|---|---|
| Archives | 1 |
| Total images | 1576 |
| Primary format | Roboflow-style YOLO vision export |
| Delivery format | Zip archives inside each folder |
Archive Inventory
| Archive | Images | Classes | Project | Source |
|---|---|---|---|---|
DA.zip |
1576 | 62 | da-w2zc8 |
dataset |
Folder Layout
archive.zip
+-- data.yaml
+-- train/
| +-- images/
| +-- labels/
+-- valid/ or val/
+-- test/
Working With the Data
- Extract the archive you want to inspect.
- Open
data.yamlto confirm class names and split locations. - Use the image and label pairs directly for YOLO training, validation, or audit workflows.
Notes
- These archives follow a Roboflow export pattern and keep the original
data.yamlmetadata inside each zip. - Several folders contain multiple related datasets; the table above lists each archive separately.