asl_aplhabet_img_classifier

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: 2.9586
• Accuracy: 0.2692

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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
No log	1.0	26	3.2666	0.0385
No log	2.0	52	3.2701	0.0385
No log	3.0	78	3.2713	0.0288
No log	4.0	104	3.2701	0.0769
No log	5.0	130	3.2584	0.0385
No log	6.0	156	3.2537	0.0577
No log	7.0	182	3.2402	0.0577
No log	8.0	208	3.2364	0.0577
No log	9.0	234	3.2055	0.0769
No log	10.0	260	3.1794	0.0769
No log	11.0	286	3.1851	0.1346
No log	12.0	312	3.1811	0.1058
No log	13.0	338	3.1594	0.1346
No log	14.0	364	3.1269	0.1635
No log	15.0	390	3.1082	0.125
No log	16.0	416	3.1019	0.2019
No log	17.0	442	3.0886	0.2019
No log	18.0	468	3.0599	0.2115
No log	19.0	494	3.0622	0.1731
3.0197	20.0	520	3.0474	0.1538
3.0197	21.0	546	3.0245	0.2115
3.0197	22.0	572	3.0386	0.1923
3.0197	23.0	598	3.0236	0.1923
3.0197	24.0	624	3.0201	0.1923
3.0197	25.0	650	3.0056	0.2212
3.0197	26.0	676	2.9649	0.25
3.0197	27.0	702	2.9900	0.2212
3.0197	28.0	728	2.9823	0.2308
3.0197	29.0	754	2.9782	0.2115
3.0197	30.0	780	3.0136	0.1635

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2