Dataset quality report
E:\Sythetic_Data\tool-detector\synthetic_raw
✓Passed
· 0 errors · 0 warnings
1500
Images
3177
Boxes
2.12
Boxes per image
0
Empty images
0
Errors
0
Warnings
Checks
| Result | Check | What it means | Found |
|---|---|---|---|
| ✓ OK | missing_label | Image has no label file | 0 |
| ✓ OK | orphan_label | Label file has no image | 0 |
| ✓ OK | unreadable_image | Image file can't be opened | 0 |
| ✓ OK | malformed_line | Label line is not 'class x y w h' with numbers | 0 |
| ✓ OK | bad_class_id | Class id is not in the class list | 0 |
| ✓ OK | out_of_bounds | Box goes outside the image (coordinates outside 0-1) | 0 |
| ✓ OK | zero_size_box | Box has zero or negative width or height | 0 |
| ✓ OK | missing_class | A class has no boxes at all | 0 |
| ✓ OK | empty_label | Image has no boxes (fine if it's a deliberate background image) | 0 |
| ✓ OK | tiny_box | Box is very small and hard to learn from | 0 |
| ✓ OK | duplicate_box | Two boxes of the same class almost completely overlap | 0 |
| ✓ OK | class_imbalance | Largest class has many more boxes than the smallest | 0 |
| ✓ OK | no_class_names | No classes.txt found, class names are guessed | 0 |
Boxes per class
Balanced classes train more evenly. Limit before a warning: 3× between largest and smallest.
Table view
| Class | Boxes | Share |
|---|---|---|
| wrench | 740 | 23.3% |
| screwdriver | 999 | 31.4% |
| hammer | 728 | 22.9% |
| pliers | 710 | 22.3% |
Boxes per image
How many objects each image holds.
Box size
Box side length as a share of the image (√ of box area).
Where boxes appear in the frame
Darker = more box centres. One hot spot in the middle means objects are always centred, which can hurt detection near the edges.
Sample images
Images with issues first (marked ⚠), then an even spread across the dataset. Resolutions: 1024x768 (902), 768x1024 (598).
All issues
No issues found.