metadata

library_name: transformers
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: emotion_classification
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.55625

emotion_classification

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.2963
Accuracy: 0.5563

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: 5e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine_with_restarts
lr_scheduler_warmup_ratio: 0.1
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0771	1.0	10	2.0698	0.1375
2.0613	2.0	20	2.0368	0.2875
2.0214	3.0	30	2.0010	0.2625
1.9314	4.0	40	1.8913	0.3
1.785	5.0	50	1.7270	0.375
1.6343	6.0	60	1.6009	0.4313
1.5327	7.0	70	1.5766	0.3937
1.452	8.0	80	1.4714	0.475
1.38	9.0	90	1.4570	0.4688
1.3061	10.0	100	1.4357	0.4688
1.2331	11.0	110	1.3691	0.4938
1.1784	12.0	120	1.3377	0.4813
1.1049	13.0	130	1.2982	0.5625
1.0938	14.0	140	1.2847	0.5188
1.0191	15.0	150	1.2630	0.575
0.9665	16.0	160	1.3427	0.4938
0.9028	17.0	170	1.3189	0.525
0.886	18.0	180	1.2599	0.5312
0.8272	19.0	190	1.3148	0.525
0.7923	20.0	200	1.2634	0.55
0.8033	21.0	210	1.2664	0.5625
0.724	22.0	220	1.2286	0.525
0.6966	23.0	230	1.3408	0.5375
0.6722	24.0	240	1.3032	0.5062
0.6816	25.0	250	1.3318	0.5062
0.6162	26.0	260	1.3775	0.4938
0.6099	27.0	270	1.2903	0.5437
0.5786	28.0	280	1.2361	0.6
0.5931	29.0	290	1.2998	0.5312
0.5849	30.0	300	1.3221	0.5062
0.5606	31.0	310	1.2756	0.5125
0.5561	32.0	320	1.3732	0.4813
0.547	33.0	330	1.3308	0.5375
0.5405	34.0	340	1.3506	0.5062
0.5419	35.0	350	1.2487	0.5625
0.5168	36.0	360	1.2269	0.525
0.5361	37.0	370	1.2993	0.55
0.5375	38.0	380	1.2806	0.575
0.5235	39.0	390	1.3404	0.5188
0.5318	40.0	400	1.3315	0.4938

Framework versions

Transformers 4.44.2
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.19.1