beit-base-patch16-224-pt22k-ft22k-finetuned-plantorgans

This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the jpodivin/plantorgans dataset. It achieves the following results on the evaluation set:

• Loss: 1.2004
• Mean Iou: 0.0260
• Mean Accuracy: 0.0325
• Overall Accuracy: 0.0435
• Accuracy Void: nan
• Accuracy Fruit: 0.1299
• Accuracy Leaf: 0.0
• Accuracy Flower: 0.0
• Accuracy Stem: 0.0001
• Iou Void: 0.0
• Iou Fruit: 0.1299
• Iou Leaf: 0.0
• Iou Flower: 0.0
• Iou Stem: 0.0001

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: 5e-05
• train_batch_size: 10
• eval_batch_size: 10
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Void	Accuracy Fruit	Accuracy Leaf	Accuracy Flower	Accuracy Stem	Iou Void	Iou Fruit	Iou Leaf	Iou Flower	Iou Stem
1.0657	1.0	575	1.1178	0.0158	0.0197	0.0261	nan	0.0776	0.0	0.0	0.0012	0.0	0.0776	0.0	0.0	0.0012
0.5398	2.0	1150	1.0816	0.0361	0.0451	0.0599	nan	0.1783	0.0000	0.0	0.0020	0.0	0.1783	0.0000	0.0	0.0020
0.4794	3.0	1725	1.2122	0.0110	0.0138	0.0179	nan	0.0517	0.0003	0.0	0.0032	0.0	0.0517	0.0003	0.0	0.0032
0.446	4.0	2300	1.3138	0.0114	0.0142	0.0189	nan	0.0562	0.0	0.0	0.0006	0.0	0.0562	0.0	0.0	0.0006
0.4422	5.0	2875	1.2360	0.0005	0.0006	0.0006	nan	0.0012	0.0	0.0	0.0013	0.0	0.0012	0.0	0.0	0.0013
0.4183	6.0	3450	1.3598	0.0177	0.0221	0.0296	nan	0.0885	0.0	0.0	0.0	0.0	0.0885	0.0	0.0	0.0
0.3921	7.0	4025	1.2523	0.0267	0.0333	0.0446	nan	0.1333	0.0000	0.0	0.0	0.0	0.1333	0.0000	0.0	0.0
0.3743	8.0	4600	1.3146	0.0401	0.0502	0.0670	nan	0.2002	0.0	0.0	0.0005	0.0	0.2001	0.0	0.0	0.0005
0.3695	9.0	5175	1.2873	0.0294	0.0368	0.0492	nan	0.1472	0.0	0.0	0.0001	0.0	0.1471	0.0	0.0	0.0001
0.3796	10.0	5750	1.2004	0.0260	0.0325	0.0435	nan	0.1299	0.0	0.0	0.0001	0.0	0.1299	0.0	0.0	0.0001

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0