update model card README.md
Browse files
README.md
CHANGED
@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
19 |
|
20 |
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on the None dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
-
- Loss: 0.
|
23 |
-
- Accuracy: 0.
|
24 |
-
- Precision: 0.
|
25 |
-
- Recall: 0.
|
26 |
-
- F1: 0.
|
27 |
|
28 |
## Model description
|
29 |
|
@@ -51,37 +51,27 @@ The following hyperparameters were used during training:
|
|
51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
- lr_scheduler_type: linear
|
53 |
- lr_scheduler_warmup_ratio: 0.1
|
54 |
-
- num_epochs:
|
55 |
|
56 |
### Training results
|
57 |
|
58 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|
59 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|
|
60 |
-
| 4.
|
61 |
-
| 3.
|
62 |
-
| 2.
|
63 |
-
| 2.
|
64 |
-
| 1.
|
65 |
-
| 1.
|
66 |
-
| 1.
|
67 |
-
| 1.
|
68 |
-
| 0.
|
69 |
-
| 0.
|
70 |
-
| 0.
|
71 |
-
| 0.
|
72 |
-
| 0.
|
73 |
-
| 0.
|
74 |
-
| 0.
|
75 |
-
| 0.4252 | 16.0 | 15104 | 0.5072 | 0.8626 | 0.8610 | 0.8626 | 0.8579 |
|
76 |
-
| 0.4239 | 17.0 | 16048 | 0.4857 | 0.8686 | 0.8678 | 0.8686 | 0.8645 |
|
77 |
-
| 0.3951 | 18.0 | 16992 | 0.4796 | 0.8695 | 0.8675 | 0.8695 | 0.8645 |
|
78 |
-
| 0.3679 | 19.0 | 17936 | 0.4685 | 0.8739 | 0.8724 | 0.8739 | 0.8695 |
|
79 |
-
| 0.3694 | 20.0 | 18880 | 0.4604 | 0.8751 | 0.8720 | 0.8751 | 0.8697 |
|
80 |
-
| 0.3435 | 21.0 | 19824 | 0.4555 | 0.8777 | 0.8755 | 0.8777 | 0.8739 |
|
81 |
-
| 0.3204 | 22.0 | 20768 | 0.4479 | 0.8783 | 0.8763 | 0.8783 | 0.8744 |
|
82 |
-
| 0.3475 | 23.0 | 21712 | 0.4433 | 0.8794 | 0.8773 | 0.8794 | 0.8753 |
|
83 |
-
| 0.338 | 24.0 | 22656 | 0.4408 | 0.8809 | 0.8785 | 0.8809 | 0.8767 |
|
84 |
-
| 0.3437 | 25.0 | 23600 | 0.4384 | 0.8814 | 0.8793 | 0.8814 | 0.8772 |
|
85 |
|
86 |
|
87 |
### Framework versions
|
|
|
19 |
|
20 |
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on the None dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.7064
|
23 |
+
- Accuracy: 0.8177
|
24 |
+
- Precision: 0.8089
|
25 |
+
- Recall: 0.8177
|
26 |
+
- F1: 0.8067
|
27 |
|
28 |
## Model description
|
29 |
|
|
|
51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
- lr_scheduler_type: linear
|
53 |
- lr_scheduler_warmup_ratio: 0.1
|
54 |
+
- num_epochs: 15
|
55 |
|
56 |
### Training results
|
57 |
|
58 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|
59 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|
|
60 |
+
| 4.6105 | 1.0 | 839 | 4.5248 | 0.1097 | 0.0579 | 0.1097 | 0.0403 |
|
61 |
+
| 3.4711 | 2.0 | 1678 | 3.3162 | 0.3000 | 0.2302 | 0.3000 | 0.2097 |
|
62 |
+
| 2.6202 | 3.0 | 2517 | 2.4445 | 0.4709 | 0.4120 | 0.4709 | 0.3939 |
|
63 |
+
| 2.0614 | 4.0 | 3356 | 1.8839 | 0.5742 | 0.5389 | 0.5742 | 0.5168 |
|
64 |
+
| 1.7026 | 5.0 | 4195 | 1.5247 | 0.6436 | 0.6180 | 0.6436 | 0.6013 |
|
65 |
+
| 1.4288 | 6.0 | 5034 | 1.2768 | 0.6979 | 0.6810 | 0.6979 | 0.6686 |
|
66 |
+
| 1.1953 | 7.0 | 5873 | 1.0960 | 0.7323 | 0.7218 | 0.7323 | 0.7077 |
|
67 |
+
| 1.058 | 8.0 | 6712 | 0.9828 | 0.7548 | 0.7441 | 0.7548 | 0.7350 |
|
68 |
+
| 0.9691 | 9.0 | 7551 | 0.9018 | 0.7718 | 0.7616 | 0.7718 | 0.7536 |
|
69 |
+
| 0.8757 | 10.0 | 8390 | 0.8380 | 0.7893 | 0.7806 | 0.7893 | 0.7756 |
|
70 |
+
| 0.8446 | 11.0 | 9229 | 0.7905 | 0.7982 | 0.7913 | 0.7982 | 0.7859 |
|
71 |
+
| 0.7711 | 12.0 | 10068 | 0.7524 | 0.8069 | 0.7995 | 0.8069 | 0.7950 |
|
72 |
+
| 0.7689 | 13.0 | 10907 | 0.7283 | 0.8123 | 0.8043 | 0.8123 | 0.8009 |
|
73 |
+
| 0.6919 | 14.0 | 11746 | 0.7133 | 0.8148 | 0.8061 | 0.8148 | 0.8036 |
|
74 |
+
| 0.694 | 15.0 | 12585 | 0.7064 | 0.8177 | 0.8089 | 0.8177 | 0.8067 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
75 |
|
76 |
|
77 |
### Framework versions
|