ALM-AHME
/

swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-Lesion-Classification-HAM10000-S

Image Classification

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

ALM-AHME commited on Jul 22, 2023

Commit

9859eea

•

1 Parent(s): b9d94a8

update model card README.md

Files changed (1) hide show

README.md +13 -18

README.md CHANGED Viewed

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft](https://huggingface.co/microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0007
-- Accuracy: 1.0
 ## Model description
@@ -46,27 +46,22 @@ 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: 0.5
-- num_epochs: 15
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.3289        | 1.0   | 114  | 1.0633          | 0.6248   |
-| 0.7956        | 2.0   | 228  | 0.5050          | 0.8103   |
-| 0.5253        | 2.99  | 342  | 0.3013          | 0.9031   |
-| 0.2958        | 4.0   | 457  | 0.1534          | 0.9524   |
-| 0.276         | 5.0   | 571  | 0.1825          | 0.9335   |
-| 0.2556        | 6.0   | 685  | 0.0723          | 0.9729   |
-| 0.3624        | 6.99  | 799  | 0.1268          | 0.9483   |
-| 0.1986        | 8.0   | 914  | 0.0522          | 0.9778   |
-| 0.1554        | 9.0   | 1028 | 0.0205          | 0.9926   |
-| 0.1636        | 10.0  | 1142 | 0.0197          | 0.9951   |
-| 0.1147        | 10.99 | 1256 | 0.0517          | 0.9836   |
-| 0.1663        | 12.0  | 1371 | 0.0056          | 0.9959   |
-| 0.094         | 13.0  | 1485 | 0.0030          | 0.9992   |
-| 0.1308        | 14.0  | 1599 | 0.0011          | 0.9992   |
-| 0.1557        | 14.97 | 1710 | 0.0007          | 1.0      |
 ### Framework versions

 This model is a fine-tuned version of [microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft](https://huggingface.co/microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.0069
+- Accuracy: 0.9975
 ## Model description
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.5
+- num_epochs: 10
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 1.2378        | 1.0   | 114  | 0.9976          | 0.5936   |
+| 0.7272        | 2.0   | 228  | 0.4749          | 0.8309   |
+| 0.4335        | 2.99  | 342  | 0.2488          | 0.9195   |
+| 0.3298        | 4.0   | 457  | 0.1700          | 0.9310   |
+| 0.177         | 5.0   | 571  | 0.2116          | 0.9261   |
+| 0.2299        | 6.0   | 685  | 0.0933          | 0.9754   |
+| 0.2586        | 6.99  | 799  | 0.0316          | 0.9869   |
+| 0.1053        | 8.0   | 914  | 0.0256          | 0.9910   |
+| 0.2159        | 9.0   | 1028 | 0.0147          | 0.9959   |
+| 0.0607        | 9.98  | 1140 | 0.0069          | 0.9975   |
 ### Framework versions