metadata

license: apache-2.0
base_model: facebook/deit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - recall
  - f1
  - precision
model-index:
  - name: deit-base-patch16-224-finetuned-ind-17-imbalanced-aadhaarmask-14687
    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: 0.8458918688803746
          - name: Recall
            type: recall
            value: 0.8458918688803746
          - name: F1
            type: f1
            value: 0.843745130911636
          - name: Precision
            type: precision
            value: 0.8521498018011563

deit-base-patch16-224-finetuned-ind-17-imbalanced-aadhaarmask-14687

This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.3635
Accuracy: 0.8459
Recall: 0.8459
F1: 0.8437
Precision: 0.8521

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
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	Recall	F1	Precision
0.6153	0.9974	293	0.6607	0.7739	0.7739	0.7506	0.7444
0.5075	1.9983	587	0.5850	0.7927	0.7927	0.7767	0.8081
0.5278	2.9991	881	0.4721	0.8199	0.8199	0.8170	0.8301
0.445	4.0	1175	0.4495	0.8186	0.8186	0.8136	0.8192
0.3781	4.9974	1468	0.4018	0.8263	0.8263	0.8249	0.8326
0.4025	5.9983	1762	0.4356	0.8221	0.8221	0.8195	0.8245
0.3409	6.9991	2056	0.3876	0.8267	0.8267	0.8248	0.8330
0.3181	8.0	2350	0.3849	0.8391	0.8391	0.8372	0.8436
0.3042	8.9974	2643	0.3850	0.8280	0.8280	0.8285	0.8347
0.2475	9.9983	2937	0.3624	0.8493	0.8493	0.8475	0.8571
0.2339	10.9991	3231	0.3865	0.8318	0.8318	0.8281	0.8307
0.2455	12.0	3525	0.3337	0.8387	0.8387	0.8371	0.8433
0.2127	12.9974	3818	0.3685	0.8306	0.8306	0.8281	0.8356
0.2288	13.9983	4112	0.3545	0.8370	0.8370	0.8352	0.8385
0.2534	14.9991	4406	0.3587	0.8429	0.8429	0.8398	0.8537
0.1911	16.0	4700	0.3573	0.8387	0.8387	0.8367	0.8396
0.2118	16.9974	4993	0.3676	0.8370	0.8370	0.8356	0.8415
0.22	17.9983	5287	0.3469	0.8357	0.8357	0.8326	0.8412
0.1938	18.9991	5581	0.3512	0.8365	0.8365	0.8343	0.8363
0.1816	19.9489	5860	0.3323	0.8463	0.8463	0.8449	0.8476

Framework versions

Transformers 4.40.1
Pytorch 2.2.0a0+81ea7a4
Datasets 2.18.0
Tokenizers 0.19.1