plant-seedlings-model-resnet-152-2

This model is a fine-tuned version of microsoft/resnet-152 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2242
• Accuracy: 0.9381

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
2.0105	0.2	100	1.8953	0.4062
0.995	0.39	200	1.0372	0.6685
0.9354	0.59	300	0.7713	0.7461
0.6444	0.79	400	0.6037	0.8026
0.6477	0.98	500	0.5981	0.7991
0.6551	1.18	600	0.5224	0.8310
0.6466	1.38	700	0.5216	0.8222
0.4006	1.57	800	0.4244	0.8541
0.4484	1.77	900	0.4513	0.8566
0.5155	1.96	1000	0.4071	0.8649
0.518	2.16	1100	0.4155	0.8679
0.3762	2.36	1200	0.4152	0.8733
0.5409	2.55	1300	0.4038	0.8728
0.3184	2.75	1400	0.3683	0.8777
0.3861	2.95	1500	0.3675	0.8811
0.4824	3.14	1600	0.4404	0.8595
0.2793	3.34	1700	0.3696	0.8816
0.4095	3.54	1800	0.3102	0.8939
0.4151	3.73	1900	0.3558	0.8875
0.4036	3.93	2000	0.3215	0.8998
0.3547	4.13	2100	0.3511	0.8885
0.3071	4.32	2200	0.3376	0.8885
0.3448	4.52	2300	0.3807	0.8743
0.3574	4.72	2400	0.2826	0.9106
0.4435	4.91	2500	0.3275	0.9013
0.2811	5.11	2600	0.3285	0.9003
0.3514	5.3	2700	0.3562	0.8949
0.2323	5.5	2800	0.3023	0.9037
0.3736	5.7	2900	0.3012	0.8998
0.2659	5.89	3000	0.3243	0.8964
0.3934	6.09	3100	0.3007	0.9042
0.1951	6.29	3200	0.2643	0.9204
0.2882	6.48	3300	0.2816	0.9175
0.1887	6.68	3400	0.2669	0.9165
0.3612	6.88	3500	0.3215	0.8993
0.1423	7.07	3600	0.2684	0.9170
0.2935	7.27	3700	0.2826	0.9072
0.1549	7.47	3800	0.2783	0.9072
0.2678	7.66	3900	0.2535	0.9140
0.1954	7.86	4000	0.2578	0.9136
0.2319	8.06	4100	0.2595	0.9106
0.2016	8.25	4200	0.2671	0.9160
0.284	8.45	4300	0.2688	0.9136
0.1635	8.64	4400	0.3101	0.9111
0.2609	8.84	4500	0.2990	0.9145
0.1826	9.04	4600	0.2630	0.9077
0.2091	9.23	4700	0.2712	0.9180
0.1217	9.43	4800	0.2550	0.9126
0.198	9.63	4900	0.2648	0.9140
0.2123	9.82	5000	0.2819	0.9116
0.1399	10.02	5100	0.2690	0.9165
0.2429	10.22	5200	0.2685	0.9194
0.1376	10.41	5300	0.2930	0.9091
0.192	10.61	5400	0.3042	0.9101
0.1872	10.81	5500	0.2693	0.9160
0.1629	11.0	5600	0.2563	0.9185
0.2487	11.2	5700	0.2476	0.9258
0.242	11.39	5800	0.2407	0.9283
0.166	11.59	5900	0.2382	0.9317
0.1181	11.79	6000	0.2576	0.9140
0.1407	11.98	6100	0.2520	0.9268
0.1931	12.18	6200	0.2634	0.9204
0.1064	12.38	6300	0.2655	0.9219
0.1261	12.57	6400	0.2569	0.9209
0.1978	12.77	6500	0.2801	0.9131
0.2031	12.97	6600	0.2541	0.9190
0.1245	13.16	6700	0.2331	0.9249
0.2824	13.36	6800	0.2573	0.9199
0.1302	13.56	6900	0.2452	0.9219
0.0825	13.75	7000	0.2384	0.9258
0.1491	13.95	7100	0.2373	0.9303
0.1859	14.15	7200	0.2623	0.9253
0.2094	14.34	7300	0.2308	0.9303
0.14	14.54	7400	0.2377	0.9298
0.1836	14.73	7500	0.2389	0.9268
0.1347	14.93	7600	0.2205	0.9327
0.0747	15.13	7700	0.2375	0.9288
0.1448	15.32	7800	0.2277	0.9342
0.0885	15.52	7900	0.2560	0.9219
0.0975	15.72	8000	0.2082	0.9293
0.1185	15.91	8100	0.2561	0.9214
0.1544	16.11	8200	0.2599	0.9283
0.0959	16.31	8300	0.2418	0.9263
0.0835	16.5	8400	0.2521	0.9352
0.0846	16.7	8500	0.2258	0.9347
0.1255	16.9	8600	0.2170	0.9342
0.1116	17.09	8700	0.2462	0.9288
0.1331	17.29	8800	0.2123	0.9420
0.0895	17.49	8900	0.2513	0.9293
0.1628	17.68	9000	0.2223	0.9283
0.2152	17.88	9100	0.2144	0.9396
0.1074	18.07	9200	0.2295	0.9376
0.1888	18.27	9300	0.2557	0.9337
0.1014	18.47	9400	0.2007	0.9411
0.0341	18.66	9500	0.2289	0.9371
0.0365	18.86	9600	0.2434	0.9337
0.1099	19.06	9700	0.2222	0.9337
0.1303	19.25	9800	0.2208	0.9317
0.1209	19.45	9900	0.2151	0.9401
0.2119	19.65	10000	0.2209	0.9376
0.0734	19.84	10100	0.2242	0.9381

Framework versions

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