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: 0.1492
Accuracy: 0.9524

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: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1633	0.992	31	1.1239	0.8571
0.4802	1.984	62	0.4213	0.9048
0.1386	2.976	93	0.2501	0.9524
0.1003	4.0	125	0.1879	0.9524
0.0871	4.992	156	0.3482	0.8571
0.0702	5.984	187	0.2048	0.9524
0.0594	6.976	218	0.2824	0.9048
0.0425	8.0	250	0.2567	0.9524
0.0398	8.992	281	0.3363	0.8571
0.0348	9.984	312	0.2518	0.9524
0.0411	10.9760	343	0.0369	1.0
0.0445	12.0	375	0.2288	0.9524
0.0353	12.992	406	0.2364	0.8571
0.0384	13.984	437	0.2255	0.9524
0.0331	14.9760	468	0.0572	1.0
0.0252	16.0	500	0.2103	0.9524
0.0337	16.992	531	0.0295	1.0
0.0302	17.984	562	0.2805	0.9048
0.0328	18.976	593	0.2127	0.9524
0.0315	19.84	620	0.1492	0.9524

Framework versions

Transformers 4.44.2
Pytorch 2.5.0+cu121
Datasets 3.1.0
Tokenizers 0.19.1

AbdoulayeDIOP/lettuce-npk-vit

lettuce-npk-vit

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

AbdoulayeDIOP/
lettuce-npk-vit