metadata

base_model: MBZUAI/swiftformer-xs
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swiftformer-xs-DMAE
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.43478260869565216

swiftformer-xs-DMAE

This model is a fine-tuned version of MBZUAI/swiftformer-xs on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.2103
Accuracy: 0.4348

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: 0.05
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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.86	3	113.8089	0.1087
No log	2.0	7	719662.6875	0.1087
113.1437	2.86	10	3930.8594	0.1087
113.1437	4.0	14	57703.1484	0.1087
113.1437	4.86	17	1127.0548	0.3261
96.1157	6.0	21	131.2882	0.1087
96.1157	6.86	24	305.9370	0.0870
96.1157	8.0	28	209.0138	0.1087
38.8121	8.86	31	30.6911	0.1304
38.8121	10.0	35	7.7503	0.4565
38.8121	10.86	38	16.1361	0.3261
10.6492	12.0	42	7.8875	0.4565
10.6492	12.86	45	4.8321	0.4565
10.6492	14.0	49	8.2543	0.4565
3.6688	14.86	52	3.3603	0.4565
3.6688	16.0	56	3.9023	0.4565
3.6688	16.86	59	3.6221	0.4565
1.5847	18.0	63	3.2802	0.4565
1.5847	18.86	66	1.8639	0.4565
1.5235	20.0	70	2.5394	0.4565
1.5235	20.86	73	1.4062	0.4783
1.5235	22.0	77	1.4145	0.3913
1.3634	22.86	80	1.3275	0.4565
1.3634	24.0	84	1.3357	0.3478
1.3634	24.86	87	1.4708	0.4565
1.2807	26.0	91	1.2909	0.4565
1.2807	26.86	94	1.3445	0.4565
1.2807	28.0	98	1.2974	0.3696
1.2625	28.86	101	1.3482	0.4565
1.2625	30.0	105	1.2866	0.3478
1.2625	30.86	108	1.2364	0.4783
1.2472	32.0	112	1.2389	0.4130
1.2472	32.86	115	1.2459	0.3478
1.2472	34.0	119	1.2090	0.4348
1.2117	34.29	120	1.2103	0.4348

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0