vGiacomov commited on
Commit
4333896
·
1 Parent(s): b5f1bfc

Update model with improved regularization and data augmentation

Browse files
README.md CHANGED
@@ -30,7 +30,7 @@ This model was trained in Google Colab using a T4 GPU and tracked with MLflow.
30
  - Bean Rust
31
  - Angular Leaf Spot
32
 
33
- **Validation Accuracy:** 0.8797
34
 
35
  ## Training Configuration
36
 
 
30
  - Bean Rust
31
  - Angular Leaf Spot
32
 
33
+ **Validation Accuracy:** 0.8722
34
 
35
  ## Training Configuration
36
 
confusion_matrix.png CHANGED
per_class_metrics.csv CHANGED
@@ -1,7 +1,7 @@
1
  ,precision,recall,f1-score,support
2
- Healthy,1.0,0.7272727272727273,0.8421052631578947,44.0
3
- Bean Rust,0.7377049180327869,1.0,0.8490566037735849,45.0
4
  Angular Leaf Spot,1.0,0.9090909090909091,0.9523809523809523,44.0
5
- accuracy,0.8796992481203008,0.8796992481203008,0.8796992481203008,0.8796992481203008
6
- macro avg,0.912568306010929,0.8787878787878788,0.8811809397708107,133.0
7
- weighted avg,0.9112535436953039,0.8796992481203008,0.8809394034099294,133.0
 
1
  ,precision,recall,f1-score,support
2
+ Healthy,1.0,0.7045454545454546,0.8266666666666667,44.0
3
+ Bean Rust,0.7258064516129032,1.0,0.8411214953271028,45.0
4
  Angular Leaf Spot,1.0,0.9090909090909091,0.9523809523809523,44.0
5
+ accuracy,0.8721804511278195,0.8721804511278195,0.8721804511278195,0.8721804511278195
6
+ macro avg,0.9086021505376344,0.8712121212121212,0.873389704791574,133.0
7
+ weighted avg,0.9072277467863206,0.8721804511278195,0.8731470866752997,133.0
resnet18_beans.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:72a98b3414f9910436d78cf79eefbe4028c8e77bb90d54248e60e51469d0829c
3
  size 44792395
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3f6bc844cbd9ce9ad9a3218d5c8cc4dbaa8079fe24296db20fb0f8b6798be1f
3
  size 44792395