Commit
•
91b14e1
1
Parent(s):
0dccd00
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,19 @@
|
|
| 1 |
---
|
| 2 |
license: openrail
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: openrail
|
| 3 |
+
tags:
|
| 4 |
+
- object-detection
|
| 5 |
+
- ultralytics
|
| 6 |
---
|
| 7 |
+
# NTU CZ3004/SC2079 Image Recognition/Symbol Detection - Week 9 - YOLOv5
|
| 8 |
+
|
| 9 |
+
CZ3004 is a module in Nanyang Technological University's Computer Science curriculum that involves creating a robot car that can navigate within an arena and around obstacles.
|
| 10 |
+
Part of the assessment is to go to obstacles and detect alphanumeric symbols pasted on them.
|
| 11 |
+
|
| 12 |
+
## Training Data
|
| 13 |
+
The training dataset had 20,000 images across 3 classes, with each class having roughly the same number of images. The images were either downloaded from RoboFlow Universe or obtained by ourselves in real life.
|
| 14 |
+
|
| 15 |
+
## Training Procedure
|
| 16 |
+
The notebook from Ultralytics was used for training. Training was done on Google Colab for 20 epochs.
|
| 17 |
+
|
| 18 |
+
## Other Models
|
| 19 |
+
There is also a [Week 8 model available](https://huggingface.co/pyesonekyaw/MDP_ImageRecognition_YOLOv5_Week_8_AY22-23_NTU-SG)
|