oscarm17 commited on
Commit
690e110
·
verified ·
1 Parent(s): 7c18b4d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -17
README.md CHANGED
@@ -13,12 +13,15 @@ https://universe.roboflow.com/sams-sgift/pest-detection-qbalv
13
  The dataset originally came with 7.3k images, but I narrowed it down to 2.5k images
14
  Each class was pretty balanced for images, with ahids being the only one not in the 400s for photos
15
  Below are the images per class:
16
- Ant: 497 images
17
- Aphid: 283 images
18
- Beetle: 413 images
19
- Caterpillar: 409 images
20
- Grasshopper: 483
21
- Weevil: 480 images
 
 
 
22
  Classes were reduced from 19 to 6 for better focus, classes were selected on how common its found in crops, and relevance to agriculture.
23
  Annotation Process:
24
  Dataset was reviewed and cleaned using roboflow, approximately reveiwed 50 images for annotations. Corrections like misidentified labels, or
@@ -41,19 +44,22 @@ Epochs: 50
41
  Image size: 640x640
42
 
43
  Evaluation Results:
44
- Overall Metrics
45
- mAP50: 0.676
46
- mAP50-95: 0.345
47
- Precision: 0.726
48
- Recall: 0.659
 
49
 
50
  Per Class Performance (Average Precision):
51
- Ant: 0.263
52
- Aphid: 0.176
53
- Beetle: 0.370
54
- Caterpillar: 0.172
55
- Grasshopper: 0.420
56
- Weevil: 0.675
 
 
57
 
58
 
59
  ![val_batch0_pred (1)](https://cdn-uploads.huggingface.co/production/uploads/69b8f1e6906a9f3f1426ea5d/8rc-64dggsCYx4AAsFdoD.jpeg)
 
13
  The dataset originally came with 7.3k images, but I narrowed it down to 2.5k images
14
  Each class was pretty balanced for images, with ahids being the only one not in the 400s for photos
15
  Below are the images per class:
16
+ | Class | Image Count |
17
+ |-------------|------------|
18
+ | Ant | 497 |
19
+ | Aphid | 283 |
20
+ | Beetle | 413 |
21
+ | Caterpillar | 409 |
22
+ | Grasshopper | 483 |
23
+ | Weevil | 480 |
24
+
25
  Classes were reduced from 19 to 6 for better focus, classes were selected on how common its found in crops, and relevance to agriculture.
26
  Annotation Process:
27
  Dataset was reviewed and cleaned using roboflow, approximately reveiwed 50 images for annotations. Corrections like misidentified labels, or
 
44
  Image size: 640x640
45
 
46
  Evaluation Results:
47
+ | Metric | Value |
48
+ |------------|-------|
49
+ | mAP50 | 0.676 |
50
+ | mAP50-95 | 0.345 |
51
+ | Precision | 0.726 |
52
+ | Recall | 0.659 |
53
 
54
  Per Class Performance (Average Precision):
55
+ | Class | Average Precision (AP) |
56
+ |-------------|------------------------|
57
+ | Ant | 0.263 |
58
+ | Aphid | 0.176 |
59
+ | Beetle | 0.370 |
60
+ | Caterpillar | 0.172 |
61
+ | Grasshopper | 0.420 |
62
+ | Weevil | 0.675 |
63
 
64
 
65
  ![val_batch0_pred (1)](https://cdn-uploads.huggingface.co/production/uploads/69b8f1e6906a9f3f1426ea5d/8rc-64dggsCYx4AAsFdoD.jpeg)