README.md

metadata

license: apache-2.0
base_model: facebook/deit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_base_adamax_00001_fold5
    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: 0.7073170731707317

hushem_1x_deit_base_adamax_00001_fold5

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: 0.6418
• Accuracy: 0.7073

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	1.3016	0.3415
1.352	2.0	12	1.2693	0.4878
1.352	3.0	18	1.2337	0.4878
1.192	4.0	24	1.1939	0.5122
1.066	5.0	30	1.1544	0.5854
1.066	6.0	36	1.1118	0.5854
0.8995	7.0	42	1.0631	0.6098
0.8995	8.0	48	1.0130	0.5854
0.7427	9.0	54	0.9666	0.6341
0.6143	10.0	60	0.9418	0.6098
0.6143	11.0	66	0.9096	0.6341
0.4971	12.0	72	0.8791	0.6341
0.4971	13.0	78	0.8576	0.6341
0.3974	14.0	84	0.8299	0.6341
0.3312	15.0	90	0.8125	0.6585
0.3312	16.0	96	0.7924	0.6585
0.2583	17.0	102	0.7878	0.6341
0.2583	18.0	108	0.7665	0.6341
0.2053	19.0	114	0.7402	0.6585
0.1711	20.0	120	0.7303	0.6585
0.1711	21.0	126	0.7219	0.6585
0.1383	22.0	132	0.7157	0.6341
0.1383	23.0	138	0.6921	0.6585
0.1073	24.0	144	0.6843	0.6585
0.0942	25.0	150	0.6833	0.6585
0.0942	26.0	156	0.6687	0.6585
0.0772	27.0	162	0.6726	0.7073
0.0772	28.0	168	0.6619	0.6585
0.0641	29.0	174	0.6481	0.6585
0.0563	30.0	180	0.6452	0.7073
0.0563	31.0	186	0.6508	0.7073
0.0487	32.0	192	0.6523	0.7073
0.0487	33.0	198	0.6468	0.7073
0.0422	34.0	204	0.6452	0.7073
0.0402	35.0	210	0.6443	0.7073
0.0402	36.0	216	0.6441	0.7073
0.0375	37.0	222	0.6438	0.7073
0.0375	38.0	228	0.6434	0.7073
0.0345	39.0	234	0.6428	0.7073
0.0342	40.0	240	0.6422	0.7073
0.0342	41.0	246	0.6417	0.7073
0.0339	42.0	252	0.6418	0.7073
0.0339	43.0	258	0.6418	0.7073
0.0327	44.0	264	0.6418	0.7073
0.0338	45.0	270	0.6418	0.7073
0.0338	46.0	276	0.6418	0.7073
0.033	47.0	282	0.6418	0.7073
0.033	48.0	288	0.6418	0.7073
0.0336	49.0	294	0.6418	0.7073
0.0342	50.0	300	0.6418	0.7073

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.15.0