plant-seedlings-model-beit-free-0-6

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

• Loss: 0.7557
• Accuracy: 0.7475

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: 0.0002
• train_batch_size: 16
• 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
2.4892	0.2	100	2.4909	0.0751
2.4906	0.39	200	2.4886	0.0756
2.3925	0.59	300	2.3344	0.1537
2.31	0.79	400	2.3306	0.1464
2.2355	0.98	500	2.2335	0.1778
2.2642	1.18	600	2.1889	0.1807
2.0806	1.38	700	2.3229	0.1680
2.1013	1.57	800	2.1519	0.2004
2.0094	1.77	900	2.0611	0.2146
2.0387	1.96	1000	2.0413	0.2210
2.0032	2.16	1100	1.9758	0.2618
1.986	2.36	1200	1.9238	0.2638
2.0885	2.55	1300	1.8944	0.2942
1.8808	2.75	1400	1.9330	0.2868
1.915	2.95	1500	1.8919	0.2814
1.958	3.14	1600	1.8762	0.3114
1.9001	3.34	1700	1.8389	0.3232
1.8572	3.54	1800	1.7978	0.3487
1.9969	3.73	1900	1.9371	0.3089
1.9186	3.93	2000	1.8055	0.3502
1.7591	4.13	2100	1.7695	0.3428
1.8368	4.32	2200	1.7498	0.3502
1.9842	4.52	2300	1.8049	0.3193
1.7606	4.72	2400	1.6730	0.3954
1.7787	4.91	2500	1.7104	0.3777
1.6377	5.11	2600	1.6647	0.3870
1.8834	5.3	2700	1.6325	0.3973
1.6149	5.5	2800	1.6722	0.3787
1.7038	5.7	2900	1.6425	0.3973
1.682	5.89	3000	1.5927	0.4180
1.6326	6.09	3100	1.4982	0.4622
1.5687	6.29	3200	1.4440	0.4774
1.3637	6.48	3300	1.4477	0.4877
1.4079	6.68	3400	1.3827	0.5020
1.3721	6.88	3500	1.4069	0.5010
1.5675	7.07	3600	1.3595	0.5083
1.5725	7.27	3700	1.3790	0.4956
1.4522	7.47	3800	1.3116	0.5378
1.4692	7.66	3900	1.3729	0.4980
1.5073	7.86	4000	1.3799	0.5216
1.2529	8.06	4100	1.2706	0.5486
1.3727	8.25	4200	1.2519	0.5535
1.2451	8.45	4300	1.2595	0.5648
1.339	8.64	4400	1.3614	0.5172
1.2858	8.84	4500	1.3028	0.5393
1.1039	9.04	4600	1.2309	0.5771
1.0351	9.23	4700	1.2678	0.5609
1.1125	9.43	4800	1.2786	0.5624
1.1667	9.63	4900	1.2131	0.5840
1.1386	9.82	5000	1.1359	0.6154
1.1888	10.02	5100	1.1309	0.6041
1.1777	10.22	5200	1.1288	0.6287
1.3693	10.41	5300	1.3827	0.5182
1.1016	10.61	5400	1.2255	0.5594
1.1527	10.81	5500	1.0772	0.6434
1.1039	11.0	5600	1.1032	0.6100
1.2502	11.2	5700	1.1230	0.6169
1.0818	11.39	5800	1.0750	0.6302
1.0872	11.59	5900	1.0397	0.6331
1.0425	11.79	6000	1.0231	0.6483
1.0791	11.98	6100	1.0250	0.6636
0.9736	12.18	6200	1.0879	0.6267
0.9788	12.38	6300	1.1334	0.5968
0.8982	12.57	6400	0.9934	0.6528
1.077	12.77	6500	0.9698	0.6812
1.0347	12.97	6600	1.0265	0.6513
0.9159	13.16	6700	0.9442	0.6788
1.1187	13.36	6800	0.9738	0.6685
0.9624	13.56	6900	1.0008	0.6699
0.922	13.75	7000	0.9502	0.6906
0.9317	13.95	7100	0.9687	0.6758
0.9979	14.15	7200	0.9869	0.6768
0.8362	14.34	7300	0.9220	0.6994
0.8449	14.54	7400	0.9181	0.6861
0.9678	14.73	7500	0.9789	0.6729
0.9119	14.93	7600	0.8879	0.7009
0.9517	15.13	7700	0.8816	0.6994
0.9688	15.32	7800	0.8803	0.7117
0.8625	15.52	7900	0.8782	0.7038
0.9121	15.72	8000	0.8225	0.7191
0.9035	15.91	8100	0.8649	0.7087
0.8762	16.11	8200	0.8427	0.7102
0.7708	16.31	8300	0.8685	0.7117
0.8893	16.5	8400	0.8178	0.7264
0.9584	16.7	8500	0.8709	0.7092
0.757	16.9	8600	0.8244	0.7254
0.8184	17.09	8700	0.8128	0.7240
0.8858	17.29	8800	0.8360	0.7156
0.7116	17.49	8900	0.7952	0.7279
0.9579	17.68	9000	0.8263	0.7274
0.7037	17.88	9100	0.7884	0.7348
1.0359	18.07	9200	0.8118	0.7402
1.067	18.27	9300	0.8203	0.7186
0.8503	18.47	9400	0.7918	0.7362
0.8552	18.66	9500	0.7972	0.7382
0.7498	18.86	9600	0.8038	0.7343
0.8542	19.06	9700	0.7799	0.7333
0.9539	19.25	9800	0.7795	0.7333
0.7369	19.45	9900	0.8103	0.7269
0.6637	19.65	10000	0.7597	0.7441
0.6712	19.84	10100	0.7557	0.7475

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3