Jazzweller
commited on
Commit
•
545448f
1
Parent(s):
4938ad0
update model card README.md
Browse files
README.md
CHANGED
@@ -29,7 +29,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
29 |
|
30 |
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 image_folder dataset.
|
31 |
It achieves the following results on the evaluation set:
|
32 |
-
- Loss: 0.
|
33 |
- Accuracy: 0.4286
|
34 |
|
35 |
## Model description
|
@@ -50,11 +50,11 @@ More information needed
|
|
50 |
|
51 |
The following hyperparameters were used during training:
|
52 |
- learning_rate: 5e-05
|
53 |
-
- train_batch_size:
|
54 |
-
- eval_batch_size:
|
55 |
- seed: 42
|
56 |
- gradient_accumulation_steps: 4
|
57 |
-
- total_train_batch_size:
|
58 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
- lr_scheduler_type: linear
|
60 |
- lr_scheduler_warmup_ratio: 0.1
|
@@ -64,9 +64,9 @@ The following hyperparameters were used during training:
|
|
64 |
|
65 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
66 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
67 |
-
| No log | 1.0 | 1 | 0.
|
68 |
-
| No log | 2.0 | 2 | 0.
|
69 |
-
| No log | 3.0 | 3 | 0.
|
70 |
|
71 |
|
72 |
### Framework versions
|
29 |
|
30 |
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 image_folder dataset.
|
31 |
It achieves the following results on the evaluation set:
|
32 |
+
- Loss: 0.9854
|
33 |
- Accuracy: 0.4286
|
34 |
|
35 |
## Model description
|
50 |
|
51 |
The following hyperparameters were used during training:
|
52 |
- learning_rate: 5e-05
|
53 |
+
- train_batch_size: 200
|
54 |
+
- eval_batch_size: 200
|
55 |
- seed: 42
|
56 |
- gradient_accumulation_steps: 4
|
57 |
+
- total_train_batch_size: 800
|
58 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
- lr_scheduler_type: linear
|
60 |
- lr_scheduler_warmup_ratio: 0.1
|
64 |
|
65 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
66 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
67 |
+
| No log | 1.0 | 1 | 0.8196 | 0.2857 |
|
68 |
+
| No log | 2.0 | 2 | 0.9640 | 0.4286 |
|
69 |
+
| No log | 3.0 | 3 | 0.9854 | 0.4286 |
|
70 |
|
71 |
|
72 |
### Framework versions
|