Training in progress, epoch 2

README.md

@@ -17,8 +17,8 @@ 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 an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7646
-- Accuracy: 0.0
+- Loss: 0.6915
+- Accuracy: 1.0
 
 ## Model description
 
@@ -37,42 +37,30 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
+- learning_rate: 3e-05
 - train_batch_size: 8
-- eval_batch_size: 8
+- eval_batch_size: 16
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
 - 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
-- mixed_precision_training: Native AMP
+- num_epochs: 16
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.3519        | 1.0   | 1    | 0.6382          | 1.0      |
-| 0.3553        | 2.0   | 2    | 0.6523          | 1.0      |
-| 0.3574        | 3.0   | 3    | 0.6724          | 1.0      |
-| 0.352         | 4.0   | 4    | 0.6655          | 1.0      |
-| 0.3405        | 5.0   | 5    | 0.6904          | 1.0      |
-| 0.3372        | 6.0   | 6    | 0.6934          | 0.0      |
-| 0.3352        | 7.0   | 7    | 0.7046          | 0.0      |
-| 0.3277        | 8.0   | 8    | 0.7319          | 0.0      |
-| 0.3285        | 9.0   | 9    | 0.7397          | 0.0      |
-| 0.3337        | 10.0  | 10   | 0.7129          | 0.0      |
-| 0.3354        | 11.0  | 11   | 0.7261          | 0.0      |
-| 0.3179        | 12.0  | 12   | 0.7651          | 0.0      |
-| 0.3199        | 13.0  | 13   | 0.7427          | 0.0      |
-| 0.3051        | 14.0  | 14   | 0.7729          | 0.0      |
-| 0.3191        | 15.0  | 15   | 0.7646          | 0.0      |
+| 0.1803        | 1.0   | 1    | 0.6093          | 1.0      |
+| 0.1831        | 2.0   | 2    | 0.6375          | 1.0      |
+| 0.1762        | 3.0   | 3    | 0.6897          | 1.0      |
+| 0.1753        | 4.0   | 4    | 0.6915          | 1.0      |
 
 
 ### Framework versions
 
-- Transformers 4.38.2
-- Pytorch 2.1.0+cu121
+- Transformers 4.38.1
+- Pytorch 2.2.1+cpu
 - Datasets 2.17.1
 - Tokenizers 0.15.2

config.json

@@ -28,5 +28,5 @@
   "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.38.2"
+  "transformers_version": "4.38.1"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:245fc7a721894ed0b5b468b75dae2989c0ec817a4e9272aca9b8c253dcdfa68c
+oid sha256:48342570b35fd6c88d9dbc474a25b71beb82f5cbc56201084cb1e41bbd66c7a0
 size 343223968

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d259fc755a8e20cd982caf031788bae046c4e0559ea31f2475c479176f95bf4
+oid sha256:48da36374a1503d4d4aff556724d41b7437cb95b8f59b5a58ae6e8ac53c74dd0
 size 4856