README.md

metadata

license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: Action_agent
    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: 0.8019047619047619

Action_agent

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.6758
• Accuracy: 0.8019

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: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1987	0.32	100	2.1640	0.3914
1.9807	0.64	200	1.9169	0.6143
1.6738	0.96	300	1.6148	0.72
1.4828	1.27	400	1.3861	0.7705
1.2768	1.59	500	1.2412	0.7590
1.1759	1.91	600	1.1169	0.7914
1.0314	2.23	700	1.0599	0.7762
0.9702	2.55	800	0.9640	0.8105
0.9559	2.87	900	0.9138	0.8076
0.858	3.18	1000	0.8605	0.8248
0.7858	3.5	1100	0.8164	0.8371
0.7898	3.82	1200	0.7917	0.8333
0.6909	4.14	1300	0.7995	0.8038
0.6619	4.46	1400	0.8194	0.7829
0.6457	4.78	1500	0.7536	0.8086
0.6155	5.1	1600	0.7212	0.8257
0.5511	5.41	1700	0.7274	0.8095
0.5486	5.73	1800	0.7048	0.8286
0.5679	6.05	1900	0.7124	0.8181
0.4914	6.37	2000	0.7277	0.8010
0.525	6.69	2100	0.6971	0.8124
0.5081	7.01	2200	0.6869	0.8162
0.5072	7.32	2300	0.6837	0.8076
0.4702	7.64	2400	0.6736	0.8152
0.4303	7.96	2500	0.6693	0.8105
0.3916	8.28	2600	0.6487	0.8238
0.4002	8.6	2700	0.6661	0.8162
0.3965	8.92	2800	0.6611	0.8143
0.3946	9.24	2900	0.6523	0.8143
0.3794	9.55	3000	0.6616	0.8048
0.3257	9.87	3100	0.6717	0.8029
0.4175	10.19	3200	0.6530	0.8057
0.3559	10.51	3300	0.6883	0.7886
0.3824	10.83	3400	0.6611	0.8
0.3589	11.15	3500	0.6659	0.8019
0.3299	11.46	3600	0.6819	0.7962
0.3736	11.78	3700	0.6405	0.8114
0.3576	12.1	3800	0.6725	0.7962
0.3454	12.42	3900	0.7025	0.7943
0.3049	12.74	4000	0.6439	0.8133
0.3363	13.06	4100	0.6352	0.8143
0.3273	13.38	4200	0.6795	0.7886
0.283	13.69	4300	0.6705	0.8
0.2607	14.01	4400	0.6732	0.7914
0.3174	14.33	4500	0.6691	0.8048
0.3189	14.65	4600	0.6602	0.8038
0.2862	14.97	4700	0.6801	0.7933
0.2895	15.29	4800	0.6579	0.8038
0.263	15.61	4900	0.6688	0.8
0.3214	15.92	5000	0.6547	0.8057
0.2867	16.24	5100	0.6775	0.7924
0.2242	16.56	5200	0.6378	0.8086
0.2839	16.88	5300	0.6761	0.7990
0.2424	17.2	5400	0.6386	0.8124
0.2666	17.52	5500	0.6493	0.8133
0.2259	17.83	5600	0.6514	0.8048
0.2533	18.15	5700	0.6676	0.8
0.2697	18.47	5800	0.6705	0.8010
0.2558	18.79	5900	0.6750	0.8076
0.2469	19.11	6000	0.6751	0.7990
0.284	19.43	6100	0.6738	0.7981
0.2534	19.75	6200	0.6758	0.8019

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2