efficientnet-b0-cocoa

This model is a fine-tuned version of google/efficientnet-b0 on the SemilleroCV/Cocoa-dataset dataset. It achieves the following results on the evaluation set:

• Loss: 0.2657
• Accuracy: 0.9097

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 1337
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 100.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0627	1.0	196	1.5223	0.5596
0.591	2.0	392	0.8975	0.8303
0.6623	3.0	588	0.6564	0.8773
0.4874	4.0	784	0.6842	0.8339
0.4671	5.0	980	0.4894	0.8809
0.5623	6.0	1176	0.4160	0.8736
0.3917	7.0	1372	0.4022	0.8845
0.3153	8.0	1568	0.4939	0.8412
0.5814	9.0	1764	0.3540	0.8773
0.5883	10.0	1960	0.3493	0.8953
0.4616	11.0	2156	0.7928	0.7762
0.499	12.0	2352	2.0659	0.2960
0.2236	13.0	2548	0.4444	0.8520
0.2083	14.0	2744	0.4640	0.8736
0.3408	15.0	2940	0.3775	0.8773
0.3529	16.0	3136	0.3519	0.8881
0.3859	17.0	3332	0.3310	0.9061
0.3557	18.0	3528	0.3475	0.8917
0.4979	19.0	3724	0.3839	0.8592
0.7133	20.0	3920	0.3032	0.9134
0.4489	21.0	4116	0.4246	0.8520
0.2605	22.0	4312	0.2951	0.8989
0.3787	23.0	4508	0.4357	0.8520
0.3015	24.0	4704	0.3990	0.8917
0.1965	25.0	4900	0.3536	0.9097
0.3903	26.0	5096	0.4166	0.8592
0.1902	27.0	5292	0.4354	0.8520
0.2089	28.0	5488	0.4089	0.8592
0.3574	29.0	5684	0.4787	0.8231
0.3532	30.0	5880	0.3165	0.9097
0.2967	31.0	6076	0.3105	0.9134
0.2364	32.0	6272	0.3560	0.9061
0.3136	33.0	6468	0.2657	0.9097
0.4061	34.0	6664	0.2680	0.9134
0.3296	35.0	6860	0.3798	0.9061
0.2905	36.0	7056	0.5098	0.8556
0.2763	37.0	7252	0.4219	0.8809
0.2454	38.0	7448	0.2852	0.9134
0.6077	39.0	7644	0.3603	0.8989
0.1966	40.0	7840	0.3519	0.8736
0.2473	41.0	8036	0.3343	0.9025
0.2795	42.0	8232	0.3384	0.9170
0.1249	43.0	8428	0.4046	0.8773
0.2943	44.0	8624	0.3953	0.8917
0.3002	45.0	8820	0.5003	0.8592
0.1525	46.0	9016	0.3232	0.9170
0.4022	47.0	9212	0.3113	0.9170
0.4994	48.0	9408	0.4494	0.8556
0.6512	49.0	9604	0.3722	0.9206
0.3152	50.0	9800	0.2852	0.9097
0.1165	51.0	9996	0.4138	0.8628
0.216	52.0	10192	0.3413	0.8953
0.1455	53.0	10388	0.3046	0.9170
0.554	54.0	10584	0.2849	0.8989
0.3586	55.0	10780	0.3517	0.9134
0.2239	56.0	10976	0.4538	0.9025
0.1725	57.0	11172	0.4492	0.8592
0.4689	58.0	11368	0.4739	0.8628
0.3565	59.0	11564	0.2831	0.9206
0.2259	60.0	11760	0.3465	0.9206
0.2212	61.0	11956	0.2884	0.9314
0.2648	62.0	12152	0.4875	0.8448
0.3438	63.0	12348	0.3989	0.9061
0.4785	64.0	12544	0.5953	0.8520
0.06	65.0	12740	0.2954	0.9278
0.1965	66.0	12936	0.5033	0.8520
0.3548	67.0	13132	0.4132	0.8809
0.1279	68.0	13328	0.3743	0.9170
0.2879	69.0	13524	0.6423	0.7762
0.1757	70.0	13720	0.5979	0.8014
0.3338	71.0	13916	0.4398	0.8989
0.1604	72.0	14112	0.5634	0.8231
0.1078	73.0	14308	0.6204	0.7762
0.258	74.0	14504	0.3685	0.8953
0.1227	75.0	14700	0.7026	0.8159
0.2257	76.0	14896	0.4048	0.9170
0.1786	77.0	15092	0.4891	0.8845
0.2006	78.0	15288	0.4216	0.8773
0.3144	79.0	15484	0.2721	0.8953
0.1969	80.0	15680	0.4270	0.8484
0.1405	81.0	15876	0.7632	0.7834
0.1427	82.0	16072	0.3249	0.9025
0.2493	83.0	16268	0.3838	0.8989
0.331	84.0	16464	0.3330	0.9206
0.1231	85.0	16660	0.3246	0.8700
0.2781	86.0	16856	0.3710	0.8736
0.7193	87.0	17052	0.3384	0.9061
0.1149	88.0	17248	0.3703	0.9097
0.0269	89.0	17444	0.5013	0.8592
0.0967	90.0	17640	0.3456	0.8989
0.177	91.0	17836	0.3799	0.8881
0.1917	92.0	18032	0.3239	0.9061
0.2082	93.0	18228	0.4861	0.8989
0.3836	94.0	18424	0.4444	0.8736
0.1	95.0	18620	0.3713	0.8845
0.1785	96.0	18816	0.4279	0.8303
0.19	97.0	19012	0.6588	0.8412
0.099	98.0	19208	0.6632	0.8267
0.1467	99.0	19404	0.4642	0.8809
0.2617	100.0	19600	0.3624	0.8809

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.21.0