plant-seedlings-model-mit

This model is a fine-tuned version of nvidia/mit-b0 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2052
• Accuracy: 0.9401

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
2.459	0.2	100	2.4084	0.1424
1.7264	0.39	200	1.5604	0.4430
1.427	0.59	300	1.2719	0.5447
1.1796	0.79	400	0.9608	0.6469
0.6449	0.98	500	0.9086	0.6783
0.819	1.18	600	0.8235	0.7230
0.711	1.38	700	0.8286	0.7161
0.6829	1.57	800	0.6853	0.7829
0.7093	1.77	900	0.8823	0.7112
0.6265	1.96	1000	0.5434	0.8129
0.6062	2.16	1100	0.4865	0.8301
0.6318	2.36	1200	0.5239	0.8256
0.5195	2.55	1300	0.5997	0.7809
0.5847	2.75	1400	0.5282	0.8099
0.4684	2.95	1500	0.4301	0.8502
0.7026	3.14	1600	0.4628	0.8522
0.443	3.34	1700	0.4201	0.8492
0.6532	3.54	1800	0.4979	0.8330
0.5021	3.73	1900	0.5098	0.8202
0.4203	3.93	2000	0.4277	0.8512
0.4201	4.13	2100	0.4046	0.8649
0.397	4.32	2200	0.5747	0.8158
0.472	4.52	2300	0.5175	0.8237
0.5614	4.72	2400	0.4351	0.8443
0.3184	4.91	2500	0.3635	0.8787
0.3409	5.11	2600	0.4374	0.8571
0.3132	5.3	2700	0.3622	0.8767
0.3928	5.5	2800	0.3522	0.8797
0.4538	5.7	2900	0.3652	0.8718
0.5516	5.89	3000	0.4128	0.8689
0.4113	6.09	3100	0.3973	0.8649
0.3365	6.29	3200	0.4116	0.8635
0.4611	6.48	3300	0.3312	0.8846
0.312	6.68	3400	0.3888	0.8679
0.3811	6.88	3500	0.3388	0.8841
0.3711	7.07	3600	0.3300	0.8954
0.4593	7.27	3700	0.3491	0.8831
0.5211	7.47	3800	0.3682	0.8895
0.2319	7.66	3900	0.3326	0.8861
0.3811	7.86	4000	0.3407	0.8910
0.4044	8.06	4100	0.3076	0.9028
0.367	8.25	4200	0.3126	0.9023
0.3862	8.45	4300	0.3281	0.8954
0.2489	8.64	4400	0.3166	0.8929
0.3197	8.84	4500	0.3564	0.8802
0.3114	9.04	4600	0.2978	0.8969
0.3589	9.23	4700	0.3438	0.8895
0.3075	9.43	4800	0.2894	0.9082
0.3862	9.63	4900	0.2880	0.9047
0.3319	9.82	5000	0.3628	0.8915
0.3022	10.02	5100	0.2624	0.9145
0.2697	10.22	5200	0.3866	0.8851
0.218	10.41	5300	0.2632	0.9101
0.3331	10.61	5400	0.3117	0.9023
0.3043	10.81	5500	0.3604	0.8900
0.3105	11.0	5600	0.2847	0.9111
0.1758	11.2	5700	0.3144	0.9082
0.2081	11.39	5800	0.2898	0.9101
0.4005	11.59	5900	0.3138	0.8998
0.264	11.79	6000	0.2792	0.9136
0.2765	11.98	6100	0.3021	0.9003
0.2595	12.18	6200	0.2625	0.9091
0.2745	12.38	6300	0.3078	0.9057
0.2437	12.57	6400	0.2533	0.9194
0.3765	12.77	6500	0.2972	0.9008
0.2911	12.97	6600	0.2909	0.9096
0.2335	13.16	6700	0.2684	0.9136
0.3099	13.36	6800	0.3057	0.9086
0.2377	13.56	6900	0.2862	0.9140
0.3159	13.75	7000	0.2271	0.9273
0.1893	13.95	7100	0.2519	0.9244
0.1703	14.15	7200	0.2616	0.9209
0.2527	14.34	7300	0.2393	0.9293
0.3772	14.54	7400	0.2662	0.9160
0.2574	14.73	7500	0.2724	0.9155
0.1803	14.93	7600	0.2549	0.9199
0.2935	15.13	7700	0.2561	0.9185
0.2105	15.32	7800	0.2202	0.9244
0.2877	15.52	7900	0.2428	0.9234
0.2467	15.72	8000	0.2531	0.9229
0.2955	15.91	8100	0.3258	0.9194
0.3136	16.11	8200	0.2430	0.9263
0.2543	16.31	8300	0.2502	0.9204
0.161	16.5	8400	0.2241	0.9352
0.194	16.7	8500	0.2313	0.9298
0.1951	16.9	8600	0.2446	0.9219
0.2515	17.09	8700	0.2476	0.9224
0.1274	17.29	8800	0.2445	0.9273
0.3035	17.49	8900	0.2704	0.9239
0.2253	17.68	9000	0.2436	0.9332
0.0982	17.88	9100	0.2523	0.9327
0.1778	18.07	9200	0.2425	0.9322
0.1362	18.27	9300	0.2653	0.9219
0.2342	18.47	9400	0.2076	0.9401
0.2231	18.66	9500	0.2238	0.9361
0.2159	18.86	9600	0.2115	0.9357
0.1826	19.06	9700	0.2079	0.9332
0.2221	19.25	9800	0.2003	0.9366
0.136	19.45	9900	0.2170	0.9401
0.0959	19.65	10000	0.1891	0.9440
0.1525	19.84	10100	0.2052	0.9401

Framework versions

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