Modelpublic

yeray142/finetune-instance-segmentation-mini-mask2former_augmentation_default_backboneFrozen

sourceHugging Faceotherupdated 2y agoView on Hugging Face

0likes7downloads

Model Card

finetune-instance-segmentation-mini-mask2formeraugmentationdefault_backboneFrozen

This model is a fine-tuned version of facebook/mask2former-swin-tiny-coco-instance on the jsalavedra/strawberry_disease dataset. It achieves the following results on the evaluation set:

—Loss: 19.5597
—Map: 0.4279
—Map 50: 0.592
—Map 75: 0.476
—Map Small: 0.3002
—Map Medium: 0.2838
—Map Large: 0.4955
—Mar 1: 0.3752
—Mar 10: 0.7188
—Mar 100: 0.7682
—Mar Small: 0.45
—Mar Medium: 0.5977
—Mar Large: 0.8285
—Map Angular leafspot: 0.162
—Mar 100 Angular leafspot: 0.7154
—Map Anthracnose fruit rot: 0.1562
—Mar 100 Anthracnose fruit rot: 0.6118
—Map Blossom blight: 0.6367
—Mar 100 Blossom blight: 0.7455
—Map Gray mold: 0.4846
—Mar 100 Gray mold: 0.7259
—Map Leaf spot: 0.7665
—Mar 100 Leaf spot: 0.9253
—Map Powdery mildew fruit: 0.1744
—Mar 100 Powdery mildew fruit: 0.8056
—Map Powdery mildew leaf: 0.6152
—Mar 100 Powdery mildew leaf: 0.8478

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: 1e-05
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 16
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: constant
—num_epochs: 10.0
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Angular leafspot	Mar 100 Angular leafspot	Map Anthracnose fruit rot	Mar 100 Anthracnose fruit rot	Map Blossom blight	Mar 100 Blossom blight	Map Gray mold	Mar 100 Gray mold	Map Leaf spot	Mar 100 Leaf spot	Map Powdery mildew fruit	Mar 100 Powdery mildew fruit	Map Powdery mildew leaf	Mar 100 Powdery mildew leaf
53.1072	1.0	91	35.8147	0.0329	0.0466	0.0372	0.0	0.0194	0.052	0.0866	0.2227	0.2877	0.0	0.1691	0.3315	0.0011	0.0577	0.0005	0.0353	0.0005	0.0455	0.1032	0.3204	0.0595	0.7537	0.0003	0.0611	0.0653	0.74
31.9378	2.0	182	31.1508	0.146	0.1929	0.1592	0.0002	0.0603	0.1944	0.1728	0.3353	0.3875	0.05	0.2558	0.456	0.0361	0.1942	0.0012	0.0353	0.0142	0.1909	0.2208	0.5352	0.4042	0.8455	0.0017	0.1222	0.3436	0.7894
28.4456	3.0	273	27.9586	0.2437	0.3525	0.2638	0.0043	0.1252	0.2879	0.2934	0.5682	0.6154	0.15	0.4384	0.695	0.0633	0.55	0.0105	0.1824	0.297	0.5682	0.3008	0.6528	0.5844	0.8805	0.0137	0.6778	0.4367	0.7961
25.1946	4.0	364	25.4498	0.2996	0.4292	0.3245	0.0868	0.1635	0.3718	0.315	0.6344	0.6812	0.45	0.5052	0.7494	0.0876	0.5827	0.0309	0.4	0.4503	0.6591	0.3582	0.6889	0.6635	0.8946	0.0214	0.7333	0.4852	0.8098
23.2545	5.0	455	23.8964	0.3353	0.4803	0.3614	0.1095	0.19	0.4087	0.3296	0.672	0.7218	0.4	0.5347	0.7931	0.1005	0.6288	0.0936	0.5235	0.5137	0.6909	0.4064	0.6972	0.675	0.9019	0.038	0.7833	0.5196	0.8267
21.9945	6.0	546	22.5007	0.366	0.5185	0.3862	0.1667	0.2226	0.4397	0.3526	0.6857	0.729	0.4	0.5301	0.8134	0.1146	0.6788	0.1253	0.5294	0.578	0.7114	0.4334	0.7	0.701	0.9082	0.0576	0.75	0.5525	0.8255
20.5361	7.0	637	21.4663	0.383	0.5312	0.4124	0.3	0.2354	0.4501	0.3424	0.6987	0.7456	0.4	0.5526	0.8183	0.1369	0.7	0.1323	0.5941	0.5833	0.7227	0.449	0.7111	0.7275	0.9097	0.0821	0.75	0.5701	0.8318
19.5394	8.0	728	20.5817	0.4007	0.5596	0.4369	0.3	0.2559	0.4733	0.3649	0.7059	0.7526	0.4	0.5733	0.8147	0.1224	0.6923	0.161	0.6059	0.5924	0.7205	0.4677	0.7204	0.7382	0.9179	0.1318	0.7722	0.5918	0.8388
18.9893	9.0	819	20.0124	0.4102	0.5697	0.4529	0.3	0.266	0.478	0.3705	0.7166	0.7613	0.4	0.5824	0.8326	0.1426	0.7	0.1322	0.6412	0.6233	0.7409	0.4696	0.7194	0.7467	0.9195	0.1517	0.7667	0.6049	0.8416
18.3583	10.0	910	19.5597	0.4279	0.592	0.476	0.3002	0.2838	0.4955	0.3752	0.7188	0.7682	0.45	0.5977	0.8285	0.162	0.7154	0.1562	0.6118	0.6367	0.7455	0.4846	0.7259	0.7665	0.9253	0.1744	0.8056	0.6152	0.8478

Framework versions

—Transformers 4.50.0.dev0
—Pytorch 2.6.0+cu124
—Datasets 3.3.2
—Tokenizers 0.21.0