README.md

metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_tiny_adamax_00001_fold4
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6666666666666666

hushem_1x_deit_tiny_adamax_00001_fold4

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

• Loss: 0.8218
• Accuracy: 0.6667

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	6	1.3850	0.3333
1.4335	2.0	12	1.3341	0.3571
1.4335	3.0	18	1.2836	0.4286
1.2369	4.0	24	1.2256	0.5238
1.1106	5.0	30	1.1743	0.4762
1.1106	6.0	36	1.1379	0.5238
0.9897	7.0	42	1.1120	0.5952
0.9897	8.0	48	1.0871	0.6190
0.869	9.0	54	1.0617	0.5952
0.7919	10.0	60	1.0389	0.5952
0.7919	11.0	66	1.0206	0.5714
0.7005	12.0	72	1.0005	0.5714
0.7005	13.0	78	0.9876	0.5714
0.6273	14.0	84	0.9709	0.5952
0.5477	15.0	90	0.9546	0.5952
0.5477	16.0	96	0.9438	0.5714
0.4708	17.0	102	0.9277	0.5952
0.4708	18.0	108	0.9166	0.6190
0.4523	19.0	114	0.9086	0.6190
0.3797	20.0	120	0.9051	0.5952
0.3797	21.0	126	0.8956	0.6190
0.3458	22.0	132	0.8852	0.6190
0.3458	23.0	138	0.8841	0.6190
0.3057	24.0	144	0.8804	0.5952
0.2867	25.0	150	0.8683	0.6429
0.2867	26.0	156	0.8580	0.6667
0.2509	27.0	162	0.8515	0.6667
0.2509	28.0	168	0.8546	0.6429
0.2322	29.0	174	0.8500	0.6667
0.2064	30.0	180	0.8396	0.6667
0.2064	31.0	186	0.8363	0.6667
0.1928	32.0	192	0.8371	0.6667
0.1928	33.0	198	0.8332	0.6667
0.1767	34.0	204	0.8261	0.6667
0.1746	35.0	210	0.8249	0.6667
0.1746	36.0	216	0.8258	0.6667
0.1557	37.0	222	0.8248	0.6667
0.1557	38.0	228	0.8243	0.6667
0.1581	39.0	234	0.8225	0.6667
0.1477	40.0	240	0.8219	0.6667
0.1477	41.0	246	0.8217	0.6667
0.149	42.0	252	0.8218	0.6667
0.149	43.0	258	0.8218	0.6667
0.1403	44.0	264	0.8218	0.6667
0.146	45.0	270	0.8218	0.6667
0.146	46.0	276	0.8218	0.6667
0.1461	47.0	282	0.8218	0.6667
0.1461	48.0	288	0.8218	0.6667
0.1422	49.0	294	0.8218	0.6667
0.1494	50.0	300	0.8218	0.6667

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1