Update README.md
Browse files
README.md
CHANGED
|
@@ -1,40 +1,14 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- biology
|
|
|
|
| 4 |
metrics:
|
| 5 |
- accuracy
|
| 6 |
pipeline_tag: image-classification
|
|
|
|
| 7 |
---
|
| 8 |
Butterfly image classification model that use pre-trained cnn model resnet18 and fine-tuned the last fully connected layer to classify 75 categories of butterfly species.
|
| 9 |
The model used the best checkpoint with 90% test accuracy. The model constructed on Pytorch environment.
|
| 10 |
|
| 11 |
-
|
| 12 |
-
Epoch:
|
| 13 |
-
Epoch: 2 Train Loss: 0.75 Train Accuracy: 0.83 Test Accuracy: 0.83
|
| 14 |
-
Epoch: 3 Train Loss: 0.60 Train Accuracy: 0.86 Test Accuracy: 0.85
|
| 15 |
-
Epoch: 4 Train Loss: 0.51 Train Accuracy: 0.88 Test Accuracy: 0.85
|
| 16 |
-
Epoch: 5 Train Loss: 0.43 Train Accuracy: 0.90 Test Accuracy: 0.86
|
| 17 |
-
Epoch: 6 Train Loss: 0.39 Train Accuracy: 0.90 Test Accuracy: 0.87
|
| 18 |
-
Epoch: 7 Train Loss: 0.35 Train Accuracy: 0.91 Test Accuracy: 0.87
|
| 19 |
-
Epoch: 8 Train Loss: 0.32 Train Accuracy: 0.92 Test Accuracy: 0.88
|
| 20 |
-
Epoch: 9 Train Loss: 0.32 Train Accuracy: 0.92 Test Accuracy: 0.89
|
| 21 |
-
Epoch: 10 Train Loss: 0.28 Train Accuracy: 0.93 Test Accuracy: 0.89
|
| 22 |
-
Epoch: 11 Train Loss: 0.28 Train Accuracy: 0.93 Test Accuracy: 0.88
|
| 23 |
-
Epoch: 12 Train Loss: 0.26 Train Accuracy: 0.94 Test Accuracy: 0.89
|
| 24 |
-
Epoch: 13 Train Loss: 0.25 Train Accuracy: 0.94 Test Accuracy: 0.89
|
| 25 |
-
Epoch: 14 Train Loss: 0.23 Train Accuracy: 0.94 Test Accuracy: 0.89
|
| 26 |
-
Epoch: 15 Train Loss: 0.23 Train Accuracy: 0.94 Test Accuracy: 0.89
|
| 27 |
-
Epoch: 16 Train Loss: 0.22 Train Accuracy: 0.94 Test Accuracy: 0.89
|
| 28 |
-
Epoch: 17 Train Loss: 0.21 Train Accuracy: 0.95 Test Accuracy: 0.89
|
| 29 |
-
Epoch: 18 Train Loss: 0.20 Train Accuracy: 0.95 Test Accuracy: 0.90
|
| 30 |
-
Epoch: 19 Train Loss: 0.20 Train Accuracy: 0.95 Test Accuracy: 0.90
|
| 31 |
-
Epoch: 20 Train Loss: 0.20 Train Accuracy: 0.95 Test Accuracy: 0.89
|
| 32 |
-
Epoch: 21 Train Loss: 0.19 Train Accuracy: 0.95 Test Accuracy: 0.90
|
| 33 |
-
Epoch: 22 Train Loss: 0.19 Train Accuracy: 0.95 Test Accuracy: 0.90
|
| 34 |
-
Epoch: 23 Train Loss: 0.19 Train Accuracy: 0.95 Test Accuracy: 0.89
|
| 35 |
-
Epoch: 24 Train Loss: 0.18 Train Accuracy: 0.95 Test Accuracy: 0.90
|
| 36 |
-
...
|
| 37 |
-
Epoch: 26 Train Loss: 0.17 Train Accuracy: 0.96 Test Accuracy: 0.90
|
| 38 |
-
Epoch: 27 Train Loss: 0.18 Train Accuracy: 0.95 Test Accuracy: 0.90
|
| 39 |
-
Epoch: 28 Train Loss: 0.17 Train Accuracy: 0.96 Test Accuracy: 0.90
|
| 40 |
-
Epoch: 29 Train Loss: 0.16 Train Accuracy: 0.96 Test Accuracy: 0.89
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- biology
|
| 4 |
+
- pytorch
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
pipeline_tag: image-classification
|
| 8 |
+
library_name: adapter-transformers
|
| 9 |
---
|
| 10 |
Butterfly image classification model that use pre-trained cnn model resnet18 and fine-tuned the last fully connected layer to classify 75 categories of butterfly species.
|
| 11 |
The model used the best checkpoint with 90% test accuracy. The model constructed on Pytorch environment.
|
| 12 |
|
| 13 |
+
training and testing result:
|
| 14 |
+
Epoch: 28 Train Loss: 0.17 Train Accuracy: 0.96 Test Accuracy: 0.90
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|