metadata

license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: activity_classification
    results: []

activity_classification

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

Loss: 0.7631
Accuracy: 0.7710

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: 3e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1235	1.0	315	1.3182	0.7099
1.0404	2.0	630	0.9831	0.7278
0.7899	3.0	945	0.9509	0.7175
0.6961	4.0	1260	0.8258	0.7460
0.615	5.0	1575	0.7890	0.7667
0.5534	6.0	1890	0.7876	0.7591
0.524	7.0	2205	0.7627	0.7663
0.4588	8.0	2520	0.8256	0.7468
0.4407	9.0	2835	0.8041	0.7615
0.4039	10.0	3150	0.8367	0.7540
0.3966	11.0	3465	0.8708	0.7492
0.366	12.0	3780	0.8410	0.7544
0.3522	13.0	4095	0.9019	0.7365
0.3495	14.0	4410	0.8240	0.7567
0.3206	15.0	4725	0.8428	0.7607
0.3172	16.0	5040	0.8626	0.7607
0.2931	17.0	5355	1.0311	0.7306
0.2943	18.0	5670	0.9393	0.7544
0.2886	19.0	5985	0.9379	0.7472
0.2785	20.0	6300	0.8911	0.7552
0.274	21.0	6615	0.9730	0.7484
0.2716	22.0	6930	0.9546	0.7504
0.2686	23.0	7245	0.8939	0.7651
0.2489	24.0	7560	0.9397	0.7480
0.257	25.0	7875	0.9298	0.7552
0.244	26.0	8190	0.9977	0.7437
0.2333	27.0	8505	0.9967	0.75
0.2376	28.0	8820	1.0012	0.7508
0.2428	29.0	9135	0.9674	0.7421
0.224	30.0	9450	1.0239	0.7361

Framework versions

Transformers 4.33.3
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3