README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: delivery_truck_classification
    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: 0.9830508474576272

delivery_truck_classification

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0375
• Accuracy: 0.9831

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: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.94	4	1.9124	0.1864
No log	1.94	8	1.8095	0.2373
No log	2.94	12	1.6757	0.3898
No log	3.94	16	1.4906	0.5254
1.8286	4.94	20	1.2704	0.6441
1.8286	5.94	24	1.0685	0.6780
1.8286	6.94	28	0.8032	0.7458
1.8286	7.94	32	0.6309	0.7627
1.8286	8.94	36	0.4989	0.8475
0.9342	9.94	40	0.4063	0.8475
0.9342	10.94	44	0.2692	0.9153
0.9342	11.94	48	0.2736	0.8983
0.9342	12.94	52	0.2116	0.9322
0.9342	13.94	56	0.1498	0.9831
0.5151	14.94	60	0.1906	0.9153
0.5151	15.94	64	0.1698	0.9492
0.5151	16.94	68	0.1432	0.9492
0.5151	17.94	72	0.1682	0.9322
0.5151	18.94	76	0.1069	0.9831
0.4009	19.94	80	0.0821	0.9831
0.4009	20.94	84	0.0903	0.9831
0.4009	21.94	88	0.1281	0.9661
0.4009	22.94	92	0.0936	0.9831
0.4009	23.94	96	0.1059	0.9661
0.3482	24.94	100	0.1431	0.9492
0.3482	25.94	104	0.0899	0.9661
0.3482	26.94	108	0.0689	0.9661
0.3482	27.94	112	0.0751	0.9661
0.3482	28.94	116	0.0891	0.9661
0.3306	29.94	120	0.0523	0.9831
0.3306	30.94	124	0.0734	0.9831
0.3306	31.94	128	0.0746	0.9831
0.3306	32.94	132	0.0474	0.9661
0.3306	33.94	136	0.0443	0.9831
0.2871	34.94	140	0.0814	0.9831
0.2871	35.94	144	0.0691	0.9831
0.2871	36.94	148	0.0531	0.9831
0.2871	37.94	152	0.0614	0.9831
0.2871	38.94	156	0.0578	0.9831
0.2754	39.94	160	0.0520	0.9831
0.2754	40.94	164	0.0537	0.9831
0.2754	41.94	168	0.0447	0.9831
0.2754	42.94	172	0.0290	1.0
0.2754	43.94	176	0.0291	1.0
0.269	44.94	180	0.0326	0.9831
0.269	45.94	184	0.0330	0.9831
0.269	46.94	188	0.0348	0.9831
0.269	47.94	192	0.0347	0.9831
0.269	48.94	196	0.0347	0.9831
0.2615	49.94	200	0.0424	0.9831
0.2615	50.94	204	0.0451	0.9831
0.2615	51.94	208	0.0433	0.9831
0.2615	52.94	212	0.0352	0.9831
0.2615	53.94	216	0.0339	0.9831
0.2386	54.94	220	0.0339	0.9831
0.2386	55.94	224	0.0339	0.9831
0.2386	56.94	228	0.0348	0.9831
0.2386	57.94	232	0.0366	0.9831
0.2386	58.94	236	0.0374	0.9831
0.2362	59.94	240	0.0375	0.9831

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2