mobilenetv2-cocoa

This model is a fine-tuned version of google/mobilenet_v2_1.0_224 on the SemilleroCV/Cocoa-dataset dataset. It achieves the following results on the evaluation set:

• Loss: 0.3226
• Accuracy: 0.8953

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
0.569	1.0	196	0.5072	0.8628
0.3973	2.0	392	0.4278	0.8700
0.5873	3.0	588	0.4138	0.8773
0.4781	4.0	784	0.4718	0.8736
0.4483	5.0	980	0.4506	0.8773
0.655	6.0	1176	0.3685	0.8953
0.3441	7.0	1372	0.4751	0.8773
0.3166	8.0	1568	0.3796	0.8809
0.5114	9.0	1764	0.4087	0.8917
0.6452	10.0	1960	0.3760	0.8989
0.4747	11.0	2156	0.4223	0.8773
0.5145	12.0	2352	1.1704	0.5957
0.1991	13.0	2548	0.3454	0.9097
0.2396	14.0	2744	0.3913	0.8700
0.3259	15.0	2940	0.3689	0.8881
0.3434	16.0	3136	0.3743	0.8736
0.389	17.0	3332	0.3657	0.9025
0.302	18.0	3528	0.4218	0.8917
0.4693	19.0	3724	0.3226	0.8953
0.6346	20.0	3920	0.3277	0.8881
0.481	21.0	4116	0.3484	0.8700
0.2628	22.0	4312	0.3942	0.9025
0.3653	23.0	4508	0.3537	0.8989
0.344	24.0	4704	0.4758	0.8809
0.2819	25.0	4900	0.4318	0.8989
0.513	26.0	5096	0.4277	0.8412
0.201	27.0	5292	0.3915	0.8953
0.2696	28.0	5488	0.4401	0.8809
0.4204	29.0	5684	0.3856	0.8953
0.316	30.0	5880	0.3576	0.8845
0.3102	31.0	6076	0.4155	0.8809
0.1489	32.0	6272	0.4147	0.8953
0.3302	33.0	6468	0.4217	0.8953
0.3271	34.0	6664	0.3321	0.9097
0.3481	35.0	6860	0.3828	0.8809
0.3329	36.0	7056	0.4045	0.8700
0.2471	37.0	7252	0.5536	0.8664
0.2007	38.0	7448	0.3503	0.8881
0.7535	39.0	7644	0.4819	0.8809
0.1851	40.0	7840	0.3762	0.8773
0.2329	41.0	8036	0.4465	0.8845
0.2889	42.0	8232	0.4696	0.9061
0.1409	43.0	8428	0.4876	0.8809
0.2683	44.0	8624	0.6134	0.8809
0.3535	45.0	8820	0.4364	0.8809
0.1683	46.0	9016	0.4059	0.8881
0.43	47.0	9212	0.3955	0.8881
0.5702	48.0	9408	0.3898	0.8809
0.8043	49.0	9604	0.5963	0.8953
0.3742	50.0	9800	0.5273	0.8989
0.1026	51.0	9996	0.3999	0.8989
0.2357	52.0	10192	0.4724	0.8592
0.2612	53.0	10388	0.4169	0.8845
0.4747	54.0	10584	0.3973	0.8917
0.4943	55.0	10780	0.5156	0.9061
0.2296	56.0	10976	0.6397	0.8917
0.1789	57.0	11172	0.5098	0.8267
0.4355	58.0	11368	0.5032	0.8917
0.3957	59.0	11564	0.4205	0.9025
0.4806	60.0	11760	0.7011	0.8917
0.2356	61.0	11956	0.7832	0.8881
0.3865	62.0	12152	0.4622	0.8917
0.3504	63.0	12348	0.5889	0.8773
0.3766	64.0	12544	0.5246	0.8592
0.1336	65.0	12740	0.6462	0.8773
0.3275	66.0	12936	0.5013	0.8628
0.3765	67.0	13132	0.4857	0.8953
0.1622	68.0	13328	0.4918	0.8845
0.2291	69.0	13524	0.5734	0.8736
0.1786	70.0	13720	0.6691	0.8231
0.3451	71.0	13916	0.7318	0.8773
0.2313	72.0	14112	0.5041	0.8700
0.1984	73.0	14308	0.6518	0.7690
0.2345	74.0	14504	0.5280	0.8845
0.0851	75.0	14700	0.6302	0.8917
0.2234	76.0	14896	0.4843	0.8809
0.2266	77.0	15092	0.4900	0.8628
0.2735	78.0	15288	0.5249	0.8736
0.2442	79.0	15484	0.5061	0.8917
0.2246	80.0	15680	0.4810	0.8664
0.3557	81.0	15876	0.6420	0.8123
0.2017	82.0	16072	0.5158	0.8845
0.249	83.0	16268	0.4364	0.9025
0.2566	84.0	16464	0.5507	0.8736
0.1012	85.0	16660	0.4728	0.8845
0.1972	86.0	16856	0.5746	0.8809
0.7922	87.0	17052	0.5262	0.8628
0.1229	88.0	17248	0.6293	0.8845
0.0248	89.0	17444	0.6193	0.8881
0.0925	90.0	17640	0.4755	0.8700
0.1968	91.0	17836	0.5528	0.8700
0.1694	92.0	18032	0.4338	0.8953
0.2083	93.0	18228	1.1286	0.8809
0.3666	94.0	18424	0.6879	0.8267
0.1358	95.0	18620	0.5071	0.8881
0.2247	96.0	18816	0.5941	0.8520
0.2682	97.0	19012	0.5219	0.8592
0.1762	98.0	19208	0.6929	0.8520
0.2368	99.0	19404	0.5324	0.8845
0.1268	100.0	19600	0.6160	0.8881

Framework versions

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