resnet50-cocoa

This model is a fine-tuned version of microsoft/resnet-50 on the SemilleroCV/Cocoa-dataset dataset. It achieves the following results on the evaluation set:

• Loss: 0.3381
• Accuracy: 0.8917

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
1.3793	1.0	196	1.4452	0.8628
0.9417	2.0	392	1.0832	0.8628
0.8546	3.0	588	0.7324	0.8628
0.6067	4.0	784	0.5761	0.8628
0.5583	5.0	980	0.5221	0.8628
0.6819	6.0	1176	0.4618	0.8628
0.4154	7.0	1372	0.4545	0.8628
0.4997	8.0	1568	0.4556	0.8628
0.6623	9.0	1764	0.4483	0.8628
0.8141	10.0	1960	0.4494	0.8628
0.5514	11.0	2156	0.4437	0.8628
0.6831	12.0	2352	0.4407	0.8664
0.2799	13.0	2548	0.4459	0.8700
0.451	14.0	2744	0.4313	0.8809
0.3901	15.0	2940	0.4340	0.8845
0.4778	16.0	3136	0.4219	0.8845
0.5531	17.0	3332	0.4304	0.8845
0.4904	18.0	3528	0.4429	0.8845
0.5398	19.0	3724	0.4144	0.8917
0.8024	20.0	3920	0.4253	0.8881
0.7022	21.0	4116	0.4232	0.8917
0.3868	22.0	4312	0.4167	0.8917
0.4075	23.0	4508	0.3917	0.8917
0.3873	24.0	4704	0.4269	0.8881
0.2382	25.0	4900	0.3913	0.8845
0.6525	26.0	5096	0.3949	0.8881
0.3207	27.0	5292	0.3967	0.8881
0.4569	28.0	5488	0.3901	0.8845
0.6184	29.0	5684	0.4114	0.8917
0.6055	30.0	5880	0.4342	0.8881
0.47	31.0	6076	0.4071	0.8917
0.3507	32.0	6272	0.3838	0.8881
0.4888	33.0	6468	0.4006	0.8881
0.4276	34.0	6664	0.3909	0.8881
0.5371	35.0	6860	0.4238	0.8917
0.4826	36.0	7056	0.3843	0.8917
0.5119	37.0	7252	0.3747	0.8845
0.4192	38.0	7448	0.4232	0.8881
1.1545	39.0	7644	0.4415	0.8881
0.3206	40.0	7840	0.3937	0.8881
0.3464	41.0	8036	0.3678	0.8881
0.4016	42.0	8232	0.3849	0.8881
0.2037	43.0	8428	0.3487	0.8881
0.3795	44.0	8624	0.4298	0.8881
0.403	45.0	8820	0.3966	0.8881
0.2754	46.0	9016	0.3785	0.8845
0.5228	47.0	9212	0.4117	0.8881
0.7263	48.0	9408	0.3726	0.8845
0.8995	49.0	9604	0.4559	0.8917
0.6844	50.0	9800	0.4164	0.8881
0.2734	51.0	9996	0.3862	0.8881
0.4179	52.0	10192	0.4386	0.8917
0.3354	53.0	10388	0.3949	0.8881
0.7031	54.0	10584	0.3910	0.8881
0.586	55.0	10780	0.4216	0.8881
0.3601	56.0	10976	0.4545	0.8881
0.362	57.0	11172	0.3760	0.8845
0.6132	58.0	11368	0.4258	0.8881
0.5605	59.0	11564	0.3972	0.8881
0.5071	60.0	11760	0.3873	0.8917
0.458	61.0	11956	0.4098	0.8881
0.4401	62.0	12152	0.3859	0.8845
0.5439	63.0	12348	0.4142	0.8917
0.6099	64.0	12544	0.3970	0.8881
0.2749	65.0	12740	0.3656	0.8809
0.581	66.0	12936	0.4203	0.8881
0.6009	67.0	13132	0.4074	0.8917
0.2388	68.0	13328	0.3594	0.8845
0.6006	69.0	13524	0.4045	0.8845
0.388	70.0	13720	0.3717	0.8881
0.552	71.0	13916	0.4239	0.8881
0.3875	72.0	14112	0.3731	0.8881
0.3105	73.0	14308	0.3434	0.8845
0.4627	74.0	14504	0.3946	0.8881
0.2931	75.0	14700	0.3950	0.8845
0.4639	76.0	14896	0.3875	0.8881
0.3534	77.0	15092	0.4009	0.8881
0.3175	78.0	15288	0.4109	0.8881
0.5334	79.0	15484	0.3918	0.8881
0.4827	80.0	15680	0.3807	0.8881
0.5162	81.0	15876	0.3624	0.8845
0.4377	82.0	16072	0.3729	0.8881
0.4487	83.0	16268	0.3981	0.8917
0.5057	84.0	16464	0.3995	0.8917
0.3421	85.0	16660	0.3554	0.8881
0.4083	86.0	16856	0.3634	0.8845
0.7634	87.0	17052	0.3970	0.8881
0.2588	88.0	17248	0.4121	0.8917
0.1584	89.0	17444	0.3711	0.8881
0.2643	90.0	17640	0.3743	0.8881
0.2771	91.0	17836	0.3726	0.8881
0.336	92.0	18032	0.3758	0.8845
0.3283	93.0	18228	0.4397	0.8917
0.7224	94.0	18424	0.3869	0.8917
0.1575	95.0	18620	0.3381	0.8917
0.4062	96.0	18816	0.3684	0.8845
0.3849	97.0	19012	0.3887	0.8881
0.2755	98.0	19208	0.3725	0.8881
0.4952	99.0	19404	0.4137	0.8917
0.3807	100.0	19600	0.3923	0.8881

Framework versions

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