Datasets:

Languages:
English
License:
BootsofLagrangian commited on
Commit
c75a4a5
·
verified ·
1 Parent(s): ca3689c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -24,9 +24,11 @@ Trained Model based on [yolov9-e](https://github.com/WongKinYiu/yolov9) is in [h
24
 
25
 
26
  # Statistics
27
-
28
  - Train: 3,991 images
29
  - Val: 399 images
 
 
 
30
 
31
  | Character Name | # in Train | # in Val | Precision | Recall | mAP50 | mAP50-95 |
32
  |:------------------:|:---:|:---:|:---:|:---:|:---:|:---:|
 
24
 
25
 
26
  # Statistics
 
27
  - Train: 3,991 images
28
  - Val: 399 images
29
+ - All labels have been manually created through human effort, with workers ethically compensated with "passion pay", came from south korea.
30
+ - Each image is named to indicate its source.
31
+
32
 
33
  | Character Name | # in Train | # in Val | Precision | Recall | mAP50 | mAP50-95 |
34
  |:------------------:|:---:|:---:|:---:|:---:|:---:|:---:|