metadata

license: apache-2.0
base_model: google/vit-base-patch16-384
tags:
  - image-classification
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: Action_model_ViT_384
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: action_class
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8611599297012302

Action_model_ViT_384

This model is a fine-tuned version of google/vit-base-patch16-384 on the action_class dataset. It achieves the following results on the evaluation set:

Loss: 0.4520
Accuracy: 0.8612

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: 8
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
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.946	0.19	100	0.7540	0.7803
0.9248	0.37	200	0.6282	0.7961
0.7968	0.56	300	0.5834	0.8102
0.6992	0.75	400	0.5647	0.8330
0.7331	0.93	500	0.5430	0.8295
0.5822	1.12	600	0.5894	0.8172
0.5906	1.31	700	0.6862	0.7909
0.5911	1.49	800	0.5369	0.8313
0.4564	1.68	900	0.4657	0.8576
0.6416	1.87	1000	0.5697	0.8190
0.5653	2.05	1100	0.6152	0.8102
0.4145	2.24	1200	0.5793	0.8225
0.4743	2.43	1300	0.4642	0.8576
0.4908	2.61	1400	0.4520	0.8612
0.523	2.8	1500	0.4989	0.8453
0.3315	2.99	1600	0.4786	0.8576
0.2779	3.17	1700	0.5546	0.8524
0.2984	3.36	1800	0.4977	0.8576
0.5914	3.54	1900	0.6296	0.8225
0.3236	3.73	2000	0.7225	0.8172
0.6194	3.92	2100	0.5783	0.8506
0.5066	4.1	2200	0.5825	0.8260
0.3532	4.29	2300	0.5606	0.8594
0.3531	4.48	2400	0.5068	0.8699
0.2573	4.66	2500	0.5632	0.8576
0.2713	4.85	2600	0.5047	0.8612
0.3538	5.04	2700	0.5988	0.8471
0.2291	5.22	2800	0.5751	0.8453
0.2976	5.41	2900	0.5781	0.8559
0.296	5.6	3000	0.5499	0.8664
0.3776	5.78	3100	0.5718	0.8612
0.2213	5.97	3200	0.5421	0.8682
0.325	6.16	3300	0.6453	0.8453
0.1594	6.34	3400	0.5558	0.8647
0.3377	6.53	3500	0.6619	0.8418
0.3743	6.72	3600	0.5446	0.8717
0.2327	6.9	3700	0.5484	0.8735
0.1659	7.09	3800	0.6629	0.8471
0.4036	7.28	3900	0.6510	0.8330
0.2084	7.46	4000	0.5640	0.8629
0.2251	7.65	4100	0.6379	0.8541
0.192	7.84	4200	0.5897	0.8629
0.1956	8.02	4300	0.5874	0.8699
0.1446	8.21	4400	0.6462	0.8594
0.2971	8.4	4500	0.5909	0.8735
0.2665	8.58	4600	0.6769	0.8612
0.2937	8.77	4700	0.6760	0.8506
0.1437	8.96	4800	0.6566	0.8489
0.1433	9.14	4900	0.6659	0.8418
0.2069	9.33	5000	0.6825	0.8541
0.2095	9.51	5100	0.6157	0.8664
0.1579	9.7	5200	0.6693	0.8629
0.1962	9.89	5300	0.6911	0.8524
0.3149	10.07	5400	0.6260	0.8559
0.2166	10.26	5500	0.6200	0.8770
0.1259	10.45	5600	0.7164	0.8576
0.1892	10.63	5700	0.7182	0.8612
0.1953	10.82	5800	0.7193	0.8418
0.2392	11.01	5900	0.6621	0.8664
0.1594	11.19	6000	0.7471	0.8489
0.2156	11.38	6100	0.7316	0.8612
0.137	11.57	6200	0.6837	0.8699
0.181	11.75	6300	0.6595	0.8647
0.2049	11.94	6400	0.6982	0.8506
0.1028	12.13	6500	0.6771	0.8682
0.1347	12.31	6600	0.6841	0.8699
0.1269	12.5	6700	0.7226	0.8594
0.2288	12.69	6800	0.7083	0.8629
0.1094	12.87	6900	0.7455	0.8471
0.0661	13.06	7000	0.7330	0.8541
0.1811	13.25	7100	0.7363	0.8436
0.2225	13.43	7200	0.7757	0.8453
0.1619	13.62	7300	0.7361	0.8576
0.2032	13.81	7400	0.7656	0.8576
0.0216	13.99	7500	0.7760	0.8629
0.2476	14.18	7600	0.7723	0.8612
0.1616	14.37	7700	0.7247	0.8787
0.1142	14.55	7800	0.7907	0.8699
0.0906	14.74	7900	0.7829	0.8647
0.2199	14.93	8000	0.7427	0.8717
0.0643	15.11	8100	0.7280	0.8699
0.1685	15.3	8200	0.8381	0.8541
0.1677	15.49	8300	0.8638	0.8506
0.1399	15.67	8400	0.8423	0.8612
0.1041	15.86	8500	0.8051	0.8541
0.2223	16.04	8600	0.7768	0.8647
0.1016	16.23	8700	0.7965	0.8647
0.065	16.42	8800	0.8331	0.8418
0.1156	16.6	8900	0.8023	0.8629
0.2263	16.79	9000	0.8116	0.8594
0.1197	16.98	9100	0.8490	0.8576
0.1931	17.16	9200	0.8194	0.8612
0.1289	17.35	9300	0.8353	0.8489
0.2039	17.54	9400	0.8163	0.8453
0.0825	17.72	9500	0.7942	0.8524
0.0712	17.91	9600	0.8027	0.8559
0.244	18.1	9700	0.7803	0.8664
0.1482	18.28	9800	0.7754	0.8629
0.1829	18.47	9900	0.7810	0.8594
0.019	18.66	10000	0.7972	0.8559
0.061	18.84	10100	0.8180	0.8576
0.117	19.03	10200	0.8319	0.8559
0.1858	19.22	10300	0.8432	0.8559
0.1087	19.4	10400	0.8273	0.8594
0.1983	19.59	10500	0.8257	0.8612
0.2453	19.78	10600	0.8177	0.8576
0.1189	19.96	10700	0.8201	0.8594

Framework versions

Transformers 4.39.3
Pytorch 2.1.2
Datasets 2.18.0
Tokenizers 0.15.2