Model save

README.md

@@ -24,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9428571428571428
+      value: 0.9047619047619048
     - name: F1
       type: f1
-      value: 0.9442260195944405
+      value: 0.9032269317983602
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -37,9 +37,9 @@ 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 imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1971
-- Accuracy: 0.9429
-- F1: 0.9442
+- Loss: 0.8131
+- Accuracy: 0.9048
+- F1: 0.9032
 
 ## Model description
 
@@ -62,7 +62,7 @@ The following hyperparameters were used during training:
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - num_epochs: 10
 
@@ -70,21 +70,21 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| 0.073         | 1.0   | 42   | 0.2416          | 0.9238   | 0.9250 |
-| 0.061         | 2.0   | 84   | 0.2160          | 0.9333   | 0.9345 |
-| 0.0543        | 3.0   | 126  | 0.2114          | 0.9429   | 0.9432 |
-| 0.0497        | 4.0   | 168  | 0.2028          | 0.9429   | 0.9442 |
-| 0.046         | 5.0   | 210  | 0.1985          | 0.9429   | 0.9442 |
-| 0.0435        | 6.0   | 252  | 0.2009          | 0.9429   | 0.9442 |
-| 0.0414        | 7.0   | 294  | 0.1976          | 0.9429   | 0.9442 |
-| 0.0402        | 8.0   | 336  | 0.1978          | 0.9429   | 0.9442 |
-| 0.0391        | 9.0   | 378  | 0.1967          | 0.9429   | 0.9442 |
-| 0.0385        | 10.0  | 420  | 0.1971          | 0.9429   | 0.9442 |
+| 2.0824        | 1.0   | 42   | 1.9709          | 0.4476   | 0.3861 |
+| 1.77          | 2.0   | 84   | 1.7102          | 0.6952   | 0.7022 |
+| 1.4691        | 3.0   | 126  | 1.4748          | 0.8476   | 0.8506 |
+| 1.198         | 4.0   | 168  | 1.2657          | 0.8952   | 0.8954 |
+| 0.9858        | 5.0   | 210  | 1.1113          | 0.9048   | 0.9035 |
+| 0.819         | 6.0   | 252  | 0.9877          | 0.9048   | 0.9032 |
+| 0.6969        | 7.0   | 294  | 0.9090          | 0.9048   | 0.9034 |
+| 0.6209        | 8.0   | 336  | 0.8524          | 0.9048   | 0.9032 |
+| 0.5651        | 9.0   | 378  | 0.8226          | 0.9048   | 0.9032 |
+| 0.5326        | 10.0  | 420  | 0.8131          | 0.9048   | 0.9032 |
 
 
 ### Framework versions
 
-- Transformers 4.45.2
-- Pytorch 2.4.1+cu121
-- Datasets 3.0.1
-- Tokenizers 0.20.0
+- Transformers 4.46.2
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

all_results.json

@@ -7,8 +7,8 @@
     "eval_samples_per_second": 37.133,
     "eval_steps_per_second": 4.951,
     "total_flos": 5.130291560557363e+17,
-    "train_loss": 0.0,
-    "train_runtime": 548.693,
-    "train_samples_per_second": 12.065,
-    "train_steps_per_second": 1.513
+    "train_loss": 1.0611230804806664,
+    "train_runtime": 556.247,
+    "train_samples_per_second": 11.901,
+    "train_steps_per_second": 0.755
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 5.130291560557363e+17,
-    "train_loss": 0.0,
-    "train_runtime": 548.693,
-    "train_samples_per_second": 12.065,
-    "train_steps_per_second": 1.513
+    "train_loss": 1.0611230804806664,
+    "train_runtime": 556.247,
+    "train_samples_per_second": 11.901,
+    "train_steps_per_second": 0.755
 }