Update README.md
Browse files
README.md
CHANGED
|
@@ -6,4 +6,15 @@ tags:
|
|
| 6 |
- brick
|
| 7 |
- bricks
|
| 8 |
---
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
- brick
|
| 7 |
- bricks
|
| 8 |
---
|
| 9 |
+
|
| 10 |
+
# Overview
|
| 11 |
+
|
| 12 |
+
The model(s) in this repository are trained with the [dreamfactor/biggest-lego-dataset-600-parts](https://www.kaggle.com/datasets/dreamfactor/biggest-lego-dataset-600-parts)
|
| 13 |
+
from Kaggle and the [Yolov7](https://github.com/WongKinYiu/yolov7) training script.
|
| 14 |
+
|
| 15 |
+
## Limitations
|
| 16 |
+
|
| 17 |
+
The `zero-shot-1000.pt` was trained in the `lego-training-1000.ipynb` notebook with 1000 images and does not differentiate lego classes but only tries to predict Lego objects.
|
| 18 |
+
This can be easily reconfigured and retrained in the notebook, but the current implementation leads to many false positives on non-Lego objects and therefore can be improved
|
| 19 |
+
upon. Also, it could be worth investigating if the metrics improve with a bigger training dataset, as currently only 1000 images are being used (approx. 0.6% of the full
|
| 20 |
+
dataset).
|