segformer-finetuned-biofilm_MRCNNv1_train

This model is a fine-tuned version of nvidia/mit-b0 on the heroza/biofilm_MRCNNv1_train dataset. It achieves the following results on the evaluation set:

• Loss: 0.0003
• Mean Iou: 0.5000
• Mean Accuracy: 1.0000
• Overall Accuracy: 1.0000
• Accuracy Background: 1.0000
• Accuracy Biofilm: nan
• Iou Background: 1.0000
• Iou Biofilm: 0.0

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: 6e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 1337
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: polynomial
• training_steps: 10000

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Biofilm	Iou Background	Iou Biofilm
0.262	1.0	136	0.0895	0.4984	0.9968	0.9968	0.9968	nan	0.9968	0.0
0.0826	2.0	272	0.0225	0.4995	0.9990	0.9990	0.9990	nan	0.9990	0.0
0.0225	3.0	408	0.0228	0.4985	0.9971	0.9971	0.9971	nan	0.9971	0.0
0.0141	4.0	544	0.0116	0.4997	0.9995	0.9995	0.9995	nan	0.9995	0.0
0.0093	5.0	680	0.0069	0.4996	0.9993	0.9993	0.9993	nan	0.9993	0.0
0.0054	6.0	816	0.0042	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.004	7.0	952	0.0030	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0034	8.0	1088	0.0027	0.4999	0.9998	0.9998	0.9998	nan	0.9998	0.0
0.0024	9.0	1224	0.0021	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0021	10.0	1360	0.0015	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0017	11.0	1496	0.0019	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0014	12.0	1632	0.0015	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0012	13.0	1768	0.0010	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0009	14.0	1904	0.0010	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0008	15.0	2040	0.0009	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0008	16.0	2176	0.0008	0.5000	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0006	17.0	2312	0.0009	0.4999	0.9999	0.9999	0.9999	nan	0.9999	0.0
0.0007	18.0	2448	0.0005	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0006	19.0	2584	0.0010	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0005	20.0	2720	0.0004	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0004	21.0	2856	0.0005	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0004	22.0	2992	0.0004	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0003	23.0	3128	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0004	24.0	3264	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0003	25.0	3400	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0002	26.0	3536	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0003	27.0	3672	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0002	28.0	3808	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0002	29.0	3944	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0002	30.0	4080	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0002	31.0	4216	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	32.0	4352	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0002	33.0	4488	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	34.0	4624	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	35.0	4760	0.0001	1.0	1.0	1.0	1.0	nan	1.0	nan
0.0002	36.0	4896	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	37.0	5032	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	38.0	5168	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	39.0	5304	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	40.0	5440	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	41.0	5576	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	42.0	5712	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	43.0	5848	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	44.0	5984	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	45.0	6120	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	46.0	6256	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	47.0	6392	0.0000	1.0	1.0	1.0	1.0	nan	1.0	nan
0.0001	48.0	6528	0.0000	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	49.0	6664	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	50.0	6800	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	51.0	6936	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	52.0	7072	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	53.0	7208	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	54.0	7344	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	55.0	7480	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	56.0	7616	0.0001	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	57.0	7752	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	58.0	7888	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	59.0	8024	0.0004	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	60.0	8160	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	61.0	8296	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	62.0	8432	0.0002	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	63.0	8568	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	64.0	8704	0.0002	1.0	1.0	1.0	1.0	nan	1.0	nan
0.0001	65.0	8840	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	66.0	8976	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	67.0	9112	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	68.0	9248	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	69.0	9384	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	70.0	9520	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	71.0	9656	0.0003	1.0	1.0	1.0	1.0	nan	1.0	nan
0.0001	72.0	9792	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	73.0	9928	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0
0.0001	73.53	10000	0.0003	0.5000	1.0000	1.0000	1.0000	nan	1.0000	0.0

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.0.0+cu117
• Datasets 2.14.4
• Tokenizers 0.15.1