metadata

license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: beit-base-patch16-224
    results: []

beit-base-patch16-224

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

Loss: 0.3752
Accuracy: 0.9388
Precision: 0.9451
Recall: 0.9388
F1 Score: 0.9412

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: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 45

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score
No log	0.9412	4	0.3599	0.8644	0.8831	0.8644	0.8152
No log	1.8824	8	0.2752	0.8983	0.8983	0.8983	0.8983
No log	2.8235	12	0.1735	0.9322	0.9293	0.9322	0.9286
0.2978	4.0	17	0.1745	0.9153	0.9311	0.9153	0.9200
0.2978	4.9412	21	0.1888	0.9153	0.9196	0.9153	0.9171
0.2978	5.8824	25	0.2819	0.8983	0.9092	0.8983	0.9024
0.2978	6.8235	29	0.5332	0.9153	0.9230	0.9153	0.9010
0.0283	8.0	34	0.5418	0.9153	0.9311	0.9153	0.9200
0.0283	8.9412	38	0.6494	0.8983	0.9092	0.8983	0.8758
0.0283	9.8824	42	0.5615	0.9153	0.9455	0.9153	0.9222
0.0061	10.8235	46	0.8767	0.8983	0.8910	0.8983	0.8857
0.0061	12.0	51	0.3859	0.9492	0.9619	0.9492	0.9520
0.0061	12.9412	55	0.4550	0.9322	0.9322	0.9322	0.9322
0.0061	13.8824	59	0.4314	0.9492	0.9477	0.9492	0.9479
0.01	14.8235	63	0.4127	0.9492	0.9619	0.9492	0.9520
0.01	16.0	68	0.3285	0.9492	0.9477	0.9492	0.9479
0.01	16.9412	72	0.3180	0.9492	0.9477	0.9492	0.9479
0.0076	17.8824	76	0.4482	0.9322	0.9293	0.9322	0.9286
0.0076	18.8235	80	0.4437	0.9322	0.9322	0.9322	0.9322
0.0076	20.0	85	0.4819	0.9322	0.9322	0.9322	0.9322
0.0076	20.9412	89	0.5133	0.9322	0.9293	0.9322	0.9286
0.0003	21.8824	93	0.4540	0.9492	0.9477	0.9492	0.9479
0.0003	22.8235	97	0.3857	0.9153	0.9196	0.9153	0.9171
0.0003	24.0	102	0.4077	0.8983	0.9092	0.8983	0.9024
0.0028	24.9412	106	0.3956	0.9492	0.9477	0.9492	0.9479
0.0028	25.8824	110	0.4671	0.9322	0.9293	0.9322	0.9286
0.0028	26.8235	114	0.3811	0.9322	0.9322	0.9322	0.9322
0.0028	28.0	119	0.3700	0.9322	0.9322	0.9322	0.9322
0.0006	28.9412	123	0.4028	0.9322	0.9322	0.9322	0.9322
0.0006	29.8824	127	0.6924	0.9153	0.9106	0.9153	0.9080
0.0006	30.8235	131	0.6949	0.9153	0.9106	0.9153	0.9080
0.0033	32.0	136	0.5889	0.9153	0.9120	0.9153	0.9132
0.0033	32.9412	140	0.5128	0.9322	0.9322	0.9322	0.9322
0.0033	33.8824	144	0.4411	0.9492	0.9522	0.9492	0.9502
0.0033	34.8235	148	0.4420	0.9492	0.9522	0.9492	0.9502
0.0013	36.0	153	0.5616	0.9322	0.9322	0.9322	0.9322
0.0013	36.9412	157	0.6365	0.9153	0.9120	0.9153	0.9132
0.0013	37.8824	161	0.6695	0.9153	0.9120	0.9153	0.9132
0.0001	38.8235	165	0.6846	0.9153	0.9120	0.9153	0.9132
0.0001	40.0	170	0.6930	0.9153	0.9120	0.9153	0.9132
0.0001	40.9412	174	0.6958	0.9153	0.9120	0.9153	0.9132
0.0001	41.8824	178	0.6967	0.9153	0.9120	0.9153	0.9132
0.0044	42.3529	180	0.6952	0.9153	0.9120	0.9153	0.9132

Framework versions

Transformers 4.40.1
Pytorch 2.2.1+cu121
Datasets 2.19.0
Tokenizers 0.19.1