Erik W
commited on
Commit
•
d80831c
1
Parent(s):
033a98d
update model card README.md
Browse files
README.md
CHANGED
@@ -24,16 +24,16 @@ model-index:
|
|
24 |
metrics:
|
25 |
- name: Accuracy
|
26 |
type: accuracy
|
27 |
-
value: 0.
|
28 |
- name: F1
|
29 |
type: f1
|
30 |
-
value: 0.
|
31 |
- name: Precision
|
32 |
type: precision
|
33 |
-
value: 0.
|
34 |
- name: Recall
|
35 |
type: recall
|
36 |
-
value: 0.
|
37 |
---
|
38 |
|
39 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -43,11 +43,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
43 |
|
44 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
45 |
It achieves the following results on the evaluation set:
|
46 |
-
- Loss: 0.
|
47 |
-
- Accuracy: 0.
|
48 |
-
- F1: 0.
|
49 |
-
- Precision: 0.
|
50 |
-
- Recall: 0.
|
51 |
|
52 |
## Model description
|
53 |
|
@@ -66,7 +66,7 @@ More information needed
|
|
66 |
### Training hyperparameters
|
67 |
|
68 |
The following hyperparameters were used during training:
|
69 |
-
- learning_rate: 0.
|
70 |
- train_batch_size: 64
|
71 |
- eval_batch_size: 64
|
72 |
- seed: 42
|
@@ -74,18 +74,18 @@ The following hyperparameters were used during training:
|
|
74 |
- total_train_batch_size: 256
|
75 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
76 |
- lr_scheduler_type: linear
|
77 |
-
- lr_scheduler_warmup_ratio: 0.
|
78 |
- num_epochs: 5
|
79 |
|
80 |
### Training results
|
81 |
|
82 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|
83 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
|
84 |
-
| 0.
|
85 |
-
| 0.
|
86 |
-
| 0.
|
87 |
-
| 0.
|
88 |
-
| 0.
|
89 |
|
90 |
|
91 |
### Framework versions
|
|
|
24 |
metrics:
|
25 |
- name: Accuracy
|
26 |
type: accuracy
|
27 |
+
value: 0.9844444444444445
|
28 |
- name: F1
|
29 |
type: f1
|
30 |
+
value: 0.9844678306487884
|
31 |
- name: Precision
|
32 |
type: precision
|
33 |
+
value: 0.9846508141836958
|
34 |
- name: Recall
|
35 |
type: recall
|
36 |
+
value: 0.9844444444444445
|
37 |
---
|
38 |
|
39 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
43 |
|
44 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
45 |
It achieves the following results on the evaluation set:
|
46 |
+
- Loss: 0.0393
|
47 |
+
- Accuracy: 0.9844
|
48 |
+
- F1: 0.9845
|
49 |
+
- Precision: 0.9847
|
50 |
+
- Recall: 0.9844
|
51 |
|
52 |
## Model description
|
53 |
|
|
|
66 |
### Training hyperparameters
|
67 |
|
68 |
The following hyperparameters were used during training:
|
69 |
+
- learning_rate: 0.0001
|
70 |
- train_batch_size: 64
|
71 |
- eval_batch_size: 64
|
72 |
- seed: 42
|
|
|
74 |
- total_train_batch_size: 256
|
75 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
76 |
- lr_scheduler_type: linear
|
77 |
+
- lr_scheduler_warmup_ratio: 0.2
|
78 |
- num_epochs: 5
|
79 |
|
80 |
### Training results
|
81 |
|
82 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|
83 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
|
84 |
+
| 0.3039 | 1.0 | 95 | 0.1300 | 0.9607 | 0.9609 | 0.9619 | 0.9607 |
|
85 |
+
| 0.2357 | 2.0 | 190 | 0.0815 | 0.9678 | 0.9678 | 0.9685 | 0.9678 |
|
86 |
+
| 0.163 | 3.0 | 285 | 0.0559 | 0.9807 | 0.9807 | 0.9809 | 0.9807 |
|
87 |
+
| 0.1267 | 4.0 | 380 | 0.0492 | 0.9837 | 0.9837 | 0.9839 | 0.9837 |
|
88 |
+
| 0.1059 | 5.0 | 475 | 0.0393 | 0.9844 | 0.9845 | 0.9847 | 0.9844 |
|
89 |
|
90 |
|
91 |
### Framework versions
|