metadata

tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - f1
model-index:
  - name: dit-base-finetuned-brs
    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: 0.5882352941176471
          - name: F1
            type: f1
            value: 0.631578947368421

dit-base-finetuned-brs

This model is a fine-tuned version of microsoft/dit-base on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 3.7504
Accuracy: 0.5882
F1: 0.6316
Precision (ppv): 0.5455
Recall (sensitivity): 0.75
Specificity: 0.4444
Npv: 0.6667
Auc: 0.5972

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: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision (ppv)	Recall (sensitivity)	Specificity	Npv	Auc
0.7296	6.25	100	0.6515	0.5294	0.5	0.5	0.5	0.5556	0.5556	0.5278
0.6136	12.49	200	0.6160	0.6471	0.5	0.75	0.375	0.8889	0.6154	0.6319
0.5701	18.74	300	0.6643	0.6471	0.5714	0.6667	0.5	0.7778	0.6364	0.6389
0.348	24.98	400	1.3046	0.5882	0.6316	0.5455	0.75	0.4444	0.6667	0.5972
0.7343	31.25	500	1.3682	0.5882	0.6316	0.5455	0.75	0.4444	0.6667	0.5972
0.4244	37.49	600	2.4365	0.5294	0.5556	0.5	0.625	0.4444	0.5714	0.5347
0.4067	43.74	700	2.1054	0.5882	0.5333	0.5714	0.5	0.6667	0.6	0.5833
0.446	49.98	800	3.2303	0.5294	0.5556	0.5	0.625	0.4444	0.5714	0.5347
0.4791	56.25	900	2.7902	0.5294	0.5	0.5	0.5	0.5556	0.5556	0.5278
0.3505	62.49	1000	2.9710	0.5882	0.5882	0.5556	0.625	0.5556	0.625	0.5903
0.0057	68.74	1100	4.3480	0.5294	0.5556	0.5	0.625	0.4444	0.5714	0.5347
0.3964	74.98	1200	3.3305	0.5294	0.5	0.5	0.5	0.5556	0.5556	0.5278
0.0253	81.25	1300	3.1798	0.5882	0.5882	0.5556	0.625	0.5556	0.625	0.5903
0.0585	87.49	1400	4.3246	0.5294	0.5556	0.5	0.625	0.4444	0.5714	0.5347
0.0917	93.74	1500	3.5914	0.5294	0.5556	0.5	0.625	0.4444	0.5714	0.5347
0.1333	99.98	1600	3.7504	0.5882	0.6316	0.5455	0.75	0.4444	0.6667	0.5972

Framework versions

Transformers 4.23.1
Pytorch 1.12.1+cu113
Datasets 2.6.1
Tokenizers 0.13.1