README.md

metadata

license: apache-2.0
base_model: facebook/deit-base-distilled-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: deit-base-distilled-patch16-224-hasta-65-fold3
    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: 0.6388888888888888

deit-base-distilled-patch16-224-hasta-65-fold3

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

• Loss: 0.7785
• Accuracy: 0.6389

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.5714	1	1.0782	0.3889
No log	1.7143	3	1.0920	0.4444
No log	2.8571	5	1.1444	0.3333
No log	4.0	7	1.0610	0.4444
No log	4.5714	8	1.0671	0.5
1.0571	5.7143	10	1.1793	0.4444
1.0571	6.8571	12	1.0866	0.4722
1.0571	8.0	14	1.1468	0.5
1.0571	8.5714	15	1.2370	0.4167
1.0571	9.7143	17	1.0559	0.5
1.0571	10.8571	19	0.9752	0.5278
0.8894	12.0	21	1.0526	0.5
0.8894	12.5714	22	1.0317	0.5556
0.8894	13.7143	24	1.0262	0.5556
0.8894	14.8571	26	0.9769	0.4722
0.8894	16.0	28	0.9297	0.4722
0.8894	16.5714	29	0.8848	0.5556
0.7239	17.7143	31	0.8514	0.5833
0.7239	18.8571	33	0.8981	0.5
0.7239	20.0	35	0.9670	0.5
0.7239	20.5714	36	0.8502	0.5556
0.7239	21.7143	38	0.7785	0.6389
0.6094	22.8571	40	0.9256	0.5556
0.6094	24.0	42	0.9037	0.5278
0.6094	24.5714	43	0.8753	0.5278
0.6094	25.7143	45	0.8113	0.5556
0.6094	26.8571	47	0.9797	0.5
0.6094	28.0	49	1.0319	0.5
0.4826	28.5714	50	0.9114	0.5
0.4826	29.7143	52	0.7637	0.6389
0.4826	30.8571	54	0.8048	0.5556
0.4826	32.0	56	0.9822	0.5
0.4826	32.5714	57	0.9031	0.5278
0.4826	33.7143	59	0.7211	0.5833
0.3943	34.8571	61	0.6979	0.6111
0.3943	36.0	63	0.7324	0.6111
0.3943	36.5714	64	0.7462	0.6389
0.3943	37.7143	66	0.7728	0.5833
0.3943	38.8571	68	0.7530	0.6389
0.3325	40.0	70	0.7361	0.6389
0.3325	40.5714	71	0.7227	0.6389
0.3325	41.7143	73	0.7938	0.5833
0.3325	42.8571	75	0.8003	0.5556
0.3325	44.0	77	0.7544	0.6389
0.3325	44.5714	78	0.7556	0.6389
0.2911	45.7143	80	0.7858	0.5833
0.2911	46.8571	82	0.7992	0.5556
0.2911	48.0	84	0.8293	0.6111
0.2911	48.5714	85	0.8294	0.5833
0.2911	49.7143	87	0.8113	0.5833
0.2911	50.8571	89	0.8062	0.5278
0.2577	52.0	91	0.8508	0.5556
0.2577	52.5714	92	0.8744	0.5556
0.2577	53.7143	94	0.8948	0.5556
0.2577	54.8571	96	0.8976	0.5278
0.2577	56.0	98	0.8933	0.5278
0.2577	56.5714	99	0.8900	0.5278
0.2642	57.1429	100	0.8886	0.5278

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1