Model save

README.md

@@ -2,7 +2,6 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
-- image-classification
 - generated_from_trainer
 datasets:
 - beans
@@ -33,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0469
+- Loss: 0.0504
 - Matthews Correlation: 0.9888
 
 ## Model description
@@ -54,7 +53,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
-- train_batch_size: 64
+- train_batch_size: 32
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -65,8 +64,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
 |:-------------:|:-----:|:----:|:---------------:|:--------------------:|
-| 0.1126        | 1.76  | 30   | 0.0969          | 0.9667               |
-| 0.031         | 3.53  | 60   | 0.0469          | 0.9888               |
+| 0.0958        | 0.91  | 30   | 0.0302          | 0.9888               |
+| 0.0289        | 1.82  | 60   | 0.0458          | 0.9888               |
+| 0.0085        | 2.73  | 90   | 0.0503          | 0.9888               |
+| 0.0065        | 3.64  | 120  | 0.0504          | 0.9888               |
 
 
 ### Framework versions

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca648f2cee80107be7cb3d2106b7b02698eb81d76e6abe82d564b07540bfed17
+oid sha256:cbb471982be87396397be4b0e1109a341772023833225fac2262d691333b47e4
 size 343272234