ThankGod
/

swin-tiny-patch4-window7-224-vit-finetuned-melanoma

Image Classification

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

ThankGod commited on Aug 21, 2023

Commit

1f0182c

•

1 Parent(s): 509d627

update model card README.md

Files changed (1) hide show

README.md +18 -13

README.md CHANGED Viewed

@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4590
-- Accuracy: 0.8384
 ## Model description
@@ -45,22 +45,27 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 10
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.2892        | 1.0   | 10   | 0.8297          | 0.7377   |
-| 0.6913        | 2.0   | 20   | 0.6707          | 0.7670   |
-| 0.5364        | 3.0   | 30   | 0.5587          | 0.7956   |
-| 0.4568        | 4.0   | 40   | 0.5270          | 0.8122   |
-| 0.4103        | 5.0   | 50   | 0.5141          | 0.8106   |
-| 0.3411        | 6.0   | 60   | 0.4820          | 0.8273   |
-| 0.2993        | 7.0   | 70   | 0.4801          | 0.8304   |
-| 0.2608        | 8.0   | 80   | 0.4643          | 0.8384   |
-| 0.2291        | 9.0   | 90   | 0.4703          | 0.8384   |
-| 0.224         | 10.0  | 100  | 0.4590          | 0.8384   |
 ### Framework versions

 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.2687
+- Accuracy: 0.9017
 ## Model description
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 15
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.5392        | 1.0   | 112  | 0.5796          | 0.7884   |
+| 0.4716        | 2.0   | 224  | 0.4380          | 0.8304   |
+| 0.3932        | 3.0   | 336  | 0.3749          | 0.8534   |
+| 0.3446        | 4.0   | 448  | 0.3851          | 0.8423   |
+| 0.3401        | 5.0   | 560  | 0.3141          | 0.8708   |
+| 0.2842        | 6.0   | 672  | 0.3309          | 0.8685   |
+| 0.3126        | 7.0   | 784  | 0.3493          | 0.8629   |
+| 0.2748        | 8.0   | 896  | 0.3391          | 0.8772   |
+| 0.2455        | 9.0   | 1008 | 0.3053          | 0.8843   |
+| 0.2361        | 10.0  | 1120 | 0.2749          | 0.8954   |
+| 0.2218        | 11.0  | 1232 | 0.2842          | 0.8994   |
+| 0.2071        | 12.0  | 1344 | 0.2507          | 0.9089   |
+| 0.2228        | 13.0  | 1456 | 0.2614          | 0.8994   |
+| 0.2018        | 14.0  | 1568 | 0.2664          | 0.9105   |
+| 0.1774        | 15.0  | 1680 | 0.2687          | 0.9017   |
 ### Framework versions