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_rms_lr0001_fold1
    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: 0.5777777777777777

hushem_1x_deit_tiny_rms_lr0001_fold1

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: 3.0724
• Accuracy: 0.5778

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: 0.0001
• 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	1.7581	0.2444
2.2176	2.0	12	1.4575	0.2444
2.2176	3.0	18	1.4070	0.2444
1.4625	4.0	24	1.4089	0.2444
1.4303	5.0	30	1.4530	0.2667
1.4303	6.0	36	1.3605	0.3778
1.3759	7.0	42	1.4068	0.3778
1.3759	8.0	48	1.3279	0.2889
1.3199	9.0	54	1.5997	0.2444
1.2335	10.0	60	1.5834	0.2444
1.2335	11.0	66	1.6144	0.3333
1.0406	12.0	72	1.1266	0.5111
1.0406	13.0	78	1.6894	0.3556
0.851	14.0	84	1.8080	0.4444
0.5856	15.0	90	2.0552	0.3778
0.5856	16.0	96	1.3379	0.4889
0.3402	17.0	102	1.4787	0.4889
0.3402	18.0	108	2.2439	0.4222
0.233	19.0	114	1.7239	0.4889
0.1016	20.0	120	2.5401	0.4222
0.1016	21.0	126	1.5433	0.5778
0.0994	22.0	132	1.8891	0.5333
0.0994	23.0	138	1.9405	0.4889
0.0839	24.0	144	1.5418	0.5778
0.0282	25.0	150	2.4010	0.5778
0.0282	26.0	156	2.6175	0.5778
0.0011	27.0	162	2.7024	0.5778
0.0011	28.0	168	2.7954	0.5778
0.0007	29.0	174	2.8362	0.5778
0.0006	30.0	180	2.8852	0.5778
0.0006	31.0	186	2.9050	0.5778
0.0005	32.0	192	2.9414	0.5778
0.0005	33.0	198	2.9746	0.5778
0.0005	34.0	204	2.9947	0.5778
0.0004	35.0	210	3.0141	0.5778
0.0004	36.0	216	3.0300	0.5778
0.0004	37.0	222	3.0447	0.5778
0.0004	38.0	228	3.0565	0.5778
0.0003	39.0	234	3.0642	0.5778
0.0003	40.0	240	3.0696	0.5778
0.0003	41.0	246	3.0717	0.5778
0.0003	42.0	252	3.0724	0.5778
0.0003	43.0	258	3.0724	0.5778
0.0003	44.0	264	3.0724	0.5778
0.0003	45.0	270	3.0724	0.5778
0.0003	46.0	276	3.0724	0.5778
0.0003	47.0	282	3.0724	0.5778
0.0003	48.0	288	3.0724	0.5778
0.0003	49.0	294	3.0724	0.5778
0.0003	50.0	300	3.0724	0.5778

Framework versions

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