Model save

README.md

@@ -0,0 +1,90 @@
+---
+license: apache-2.0
+base_model: google/vit-base-patch16-224
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+model-index:
+- name: vit-epsilon-1e-7
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# vit-epsilon-1e-7
+
+This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6523
+- Accuracy: 0.8714
+- Precision: 0.8675
+- Recall: 0.8714
+- F1: 0.8677
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 16
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.99) and epsilon=1e-07
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 1733
+- num_epochs: 100
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
+| 1.765         | 1.0   | 321  | 0.9570          | 0.6917   | 0.6487    | 0.6917 | 0.6531 |
+| 1.1815        | 2.0   | 642  | 0.6949          | 0.7559   | 0.7315    | 0.7559 | 0.7314 |
+| 1.0605        | 3.0   | 963  | 0.6213          | 0.7715   | 0.7649    | 0.7715 | 0.7530 |
+| 1.0147        | 4.0   | 1284 | 0.5768          | 0.7732   | 0.7807    | 0.7732 | 0.7734 |
+| 0.93          | 5.0   | 1605 | 0.6572          | 0.7587   | 0.7940    | 0.7587 | 0.7662 |
+| 0.9793        | 6.0   | 1926 | 0.6165          | 0.7701   | 0.7940    | 0.7701 | 0.7742 |
+| 0.8662        | 7.0   | 2247 | 0.6535          | 0.7240   | 0.8098    | 0.7240 | 0.7456 |
+| 0.7767        | 8.0   | 2568 | 0.5813          | 0.7566   | 0.8124    | 0.7566 | 0.7733 |
+| 0.7572        | 9.0   | 2889 | 0.5385          | 0.8145   | 0.8131    | 0.8145 | 0.8114 |
+| 0.7003        | 10.0  | 3210 | 0.5355          | 0.8027   | 0.8276    | 0.8027 | 0.8093 |
+| 0.6316        | 11.0  | 3531 | 0.6285          | 0.7653   | 0.8322    | 0.7653 | 0.7816 |
+| 0.5723        | 12.0  | 3852 | 0.5775          | 0.8017   | 0.8279    | 0.8017 | 0.8105 |
+| 0.4717        | 13.0  | 4173 | 0.5348          | 0.8350   | 0.8391    | 0.8350 | 0.8350 |
+| 0.4472        | 14.0  | 4494 | 0.5469          | 0.8239   | 0.8442    | 0.8239 | 0.8299 |
+| 0.3764        | 15.0  | 4815 | 0.5738          | 0.8291   | 0.8501    | 0.8291 | 0.8355 |
+| 0.3346        | 16.0  | 5136 | 0.5368          | 0.8436   | 0.8512    | 0.8436 | 0.8461 |
+| 0.2826        | 17.0  | 5457 | 0.5510          | 0.8474   | 0.8489    | 0.8474 | 0.8468 |
+| 0.2659        | 18.0  | 5778 | 0.5467          | 0.8547   | 0.8560    | 0.8547 | 0.8549 |
+| 0.2545        | 19.0  | 6099 | 0.6156          | 0.8433   | 0.8617    | 0.8433 | 0.8487 |
+| 0.2123        | 20.0  | 6420 | 0.6871          | 0.8429   | 0.8499    | 0.8429 | 0.8427 |
+| 0.1655        | 21.0  | 6741 | 0.6139          | 0.8610   | 0.8552    | 0.8610 | 0.8567 |
+| 0.1246        | 22.0  | 7062 | 0.6129          | 0.8675   | 0.8681    | 0.8675 | 0.8677 |
+| 0.1394        | 23.0  | 7383 | 0.6523          | 0.8714   | 0.8675    | 0.8714 | 0.8677 |
+
+
+### Framework versions
+
+- Transformers 4.40.0.dev0
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:945addb3ccc36a4a1e2cc042d71484cc36b0c1a83fc156f42ac965fdc85cdd05
+oid sha256:2808f6165151ed9c62538b0df5ae6e867ddfb573ecfdb2ec119129734ba1bd3f
 size 343239356