You can deploy this model and start earning money today!

MedicalVision/
detr_balloon_example

Name: detr_balloon_example
Rating: 2.5 (16 reviews)

Views

No views yet

Original result

IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.018
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.028
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.018
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.061
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.031
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.040
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.136
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.468
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.393
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.557
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = -1.000

After training result

IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.581
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.740
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.661
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.580
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.722
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.216
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.686
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.704
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.615
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.809
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = -1.000

Config

dataset: NIH
original model: facebook/detr-resnet-50
lr: 5e-06
dropout_rate: 0.1
weight_decay: 0.05
max_epochs: 30
train samples: 61

Logging

Training process

{'validation_loss': tensor(2.3278, device='cuda:0'), 'validation_loss_ce': tensor(0.8063, device='cuda:0'), 'validation_loss_bbox': tensor(0.1393, device='cuda:0'), 'validation_loss_giou': tensor(0.4124, device='cuda:0'), 'validation_cardinality_error': tensor(95.7500, device='cuda:0')}
{'training_loss': tensor(0.9300, device='cuda:0'), 'train_loss_ce': tensor(0.8402, device='cuda:0'), 'train_loss_bbox': tensor(0.0106, device='cuda:0'), 'train_loss_giou': tensor(0.0183, device='cuda:0'), 'train_cardinality_error': tensor(99., device='cuda:0'), 'validation_loss': tensor(1.6376, device='cuda:0'), 'validation_loss_ce': tensor(0.7839, device='cuda:0'), 'validation_loss_bbox': tensor(0.0598, device='cuda:0'), 'validation_loss_giou': tensor(0.2774, device='cuda:0'), 'validation_cardinality_error': tensor(92.4615, device='cuda:0')}
{'training_loss': tensor(0.8859, device='cuda:0'), 'train_loss_ce': tensor(0.7299, device='cuda:0'), 'train_loss_bbox': tensor(0.0061, device='cuda:0'), 'train_loss_giou': tensor(0.0626, device='cuda:0'), 'train_cardinality_error': tensor(85., device='cuda:0'), 'validation_loss': tensor(1.4878, device='cuda:0'), 'validation_loss_ce': tensor(0.7549, device='cuda:0'), 'validation_loss_bbox': tensor(0.0428, device='cuda:0'), 'validation_loss_giou': tensor(0.2594, device='cuda:0'), 'validation_cardinality_error': tensor(87.4615, device='cuda:0')}
{'training_loss': tensor(0.9309, device='cuda:0'), 'train_loss_ce': tensor(0.6880, device='cuda:0'), 'train_loss_bbox': tensor(0.0039, device='cuda:0'), 'train_loss_giou': tensor(0.1115, device='cuda:0'), 'train_cardinality_error': tensor(37., device='cuda:0'), 'validation_loss': tensor(1.4015, device='cuda:0'), 'validation_loss_ce': tensor(0.7223, device='cuda:0'), 'validation_loss_bbox': tensor(0.0418, device='cuda:0'), 'validation_loss_giou': tensor(0.2352, device='cuda:0'), 'validation_cardinality_error': tensor(75.2308, device='cuda:0')}
{'training_loss': tensor(0.9357, device='cuda:0'), 'train_loss_ce': tensor(0.6722, device='cuda:0'), 'train_loss_bbox': tensor(0.0161, device='cuda:0'), 'train_loss_giou': tensor(0.0915, device='cuda:0'), 'train_cardinality_error': tensor(50., device='cuda:0'), 'validation_loss': tensor(1.2880, device='cuda:0'), 'validation_loss_ce': tensor(0.6971, device='cuda:0'), 'validation_loss_bbox': tensor(0.0326, device='cuda:0'), 'validation_loss_giou': tensor(0.2139, device='cuda:0'), 'validation_cardinality_error': tensor(61.9231, device='cuda:0')}
{'training_loss': tensor(1.0329, device='cuda:0'), 'train_loss_ce': tensor(0.6268, device='cuda:0'), 'train_loss_bbox': tensor(0.0259, device='cuda:0'), 'train_loss_giou': tensor(0.1382, device='cuda:0'), 'train_cardinality_error': tensor(60., device='cuda:0'), 'validation_loss': tensor(1.3270, device='cuda:0'), 'validation_loss_ce': tensor(0.6700, device='cuda:0'), 'validation_loss_bbox': tensor(0.0406, device='cuda:0'), 'validation_loss_giou': tensor(0.2270, device='cuda:0'), 'validation_cardinality_error': tensor(44., device='cuda:0')}
{'training_loss': tensor(0.6385, device='cuda:0'), 'train_loss_ce': tensor(0.6109, device='cuda:0'), 'train_loss_bbox': tensor(0.0034, device='cuda:0'), 'train_loss_giou': tensor(0.0053, device='cuda:0'), 'train_cardinality_error': tensor(16., device='cuda:0'), 'validation_loss': tensor(1.2555, device='cuda:0'), 'validation_loss_ce': tensor(0.6517, device='cuda:0'), 'validation_loss_bbox': tensor(0.0333, device='cuda:0'), 'validation_loss_giou': tensor(0.2188, device='cuda:0'), 'validation_cardinality_error': tensor(35.3846, device='cuda:0')}
{'training_loss': tensor(0.9198, device='cuda:0'), 'train_loss_ce': tensor(0.6531, device='cuda:0'), 'train_loss_bbox': tensor(0.0121, device='cuda:0'), 'train_loss_giou': tensor(0.1032, device='cuda:0'), 'train_cardinality_error': tensor(48., device='cuda:0'), 'validation_loss': tensor(1.1704, device='cuda:0'), 'validation_loss_ce': tensor(0.6298, device='cuda:0'), 'validation_loss_bbox': tensor(0.0305, device='cuda:0'), 'validation_loss_giou': tensor(0.1940, device='cuda:0'), 'validation_cardinality_error': tensor(23.8462, device='cuda:0')}
{'training_loss': tensor(1.2434, device='cuda:0'), 'train_loss_ce': tensor(0.6117, device='cuda:0'), 'train_loss_bbox': tensor(0.0905, device='cuda:0'), 'train_loss_giou': tensor(0.0897, device='cuda:0'), 'train_cardinality_error': tensor(14., device='cuda:0'), 'validation_loss': tensor(1.1164, device='cuda:0'), 'validation_loss_ce': tensor(0.6053, device='cuda:0'), 'validation_loss_bbox': tensor(0.0269, device='cuda:0'), 'validation_loss_giou': tensor(0.1882, device='cuda:0'), 'validation_cardinality_error': tensor(16.9231, device='cuda:0')}
{'training_loss': tensor(0.6062, device='cuda:0'), 'train_loss_ce': tensor(0.5261, device='cuda:0'), 'train_loss_bbox': tensor(0.0094, device='cuda:0'), 'train_loss_giou': tensor(0.0166, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.0347, device='cuda:0'), 'validation_loss_ce': tensor(0.5863, device='cuda:0'), 'validation_loss_bbox': tensor(0.0253, device='cuda:0'), 'validation_loss_giou': tensor(0.1610, device='cuda:0'), 'validation_cardinality_error': tensor(12.1538, device='cuda:0')}
{'training_loss': tensor(0.9271, device='cuda:0'), 'train_loss_ce': tensor(0.5947, device='cuda:0'), 'train_loss_bbox': tensor(0.0129, device='cuda:0'), 'train_loss_giou': tensor(0.1340, device='cuda:0'), 'train_cardinality_error': tensor(5., device='cuda:0'), 'validation_loss': tensor(1.0357, device='cuda:0'), 'validation_loss_ce': tensor(0.5681, device='cuda:0'), 'validation_loss_bbox': tensor(0.0275, device='cuda:0'), 'validation_loss_giou': tensor(0.1649, device='cuda:0'), 'validation_cardinality_error': tensor(9.7692, device='cuda:0')}
{'training_loss': tensor(0.5119, device='cuda:0'), 'train_loss_ce': tensor(0.4507, device='cuda:0'), 'train_loss_bbox': tensor(0.0033, device='cuda:0'), 'train_loss_giou': tensor(0.0225, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.0390, device='cuda:0'), 'validation_loss_ce': tensor(0.5564, device='cuda:0'), 'validation_loss_bbox': tensor(0.0279, device='cuda:0'), 'validation_loss_giou': tensor(0.1716, device='cuda:0'), 'validation_cardinality_error': tensor(8.4615, device='cuda:0')}
{'training_loss': tensor(0.5317, device='cuda:0'), 'train_loss_ce': tensor(0.4902, device='cuda:0'), 'train_loss_bbox': tensor(0.0038, device='cuda:0'), 'train_loss_giou': tensor(0.0112, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.1582, device='cuda:0'), 'validation_loss_ce': tensor(0.5454, device='cuda:0'), 'validation_loss_bbox': tensor(0.0349, device='cuda:0'), 'validation_loss_giou': tensor(0.2191, device='cuda:0'), 'validation_cardinality_error': tensor(7.7692, device='cuda:0')}
{'training_loss': tensor(0.8191, device='cuda:0'), 'train_loss_ce': tensor(0.5887, device='cuda:0'), 'train_loss_bbox': tensor(0.0254, device='cuda:0'), 'train_loss_giou': tensor(0.0517, device='cuda:0'), 'train_cardinality_error': tensor(4., device='cuda:0'), 'validation_loss': tensor(1.1191, device='cuda:0'), 'validation_loss_ce': tensor(0.5417, device='cuda:0'), 'validation_loss_bbox': tensor(0.0324, device='cuda:0'), 'validation_loss_giou': tensor(0.2078, device='cuda:0'), 'validation_cardinality_error': tensor(6.2308, device='cuda:0')}
{'training_loss': tensor(0.5972, device='cuda:0'), 'train_loss_ce': tensor(0.5358, device='cuda:0'), 'train_loss_bbox': tensor(0.0051, device='cuda:0'), 'train_loss_giou': tensor(0.0180, device='cuda:0'), 'train_cardinality_error': tensor(2., device='cuda:0'), 'validation_loss': tensor(0.9988, device='cuda:0'), 'validation_loss_ce': tensor(0.5194, device='cuda:0'), 'validation_loss_bbox': tensor(0.0263, device='cuda:0'), 'validation_loss_giou': tensor(0.1740, device='cuda:0'), 'validation_cardinality_error': tensor(5.6923, device='cuda:0')}
{'training_loss': tensor(0.6085, device='cuda:0'), 'train_loss_ce': tensor(0.5220, device='cuda:0'), 'train_loss_bbox': tensor(0.0068, device='cuda:0'), 'train_loss_giou': tensor(0.0262, device='cuda:0'), 'train_cardinality_error': tensor(2., device='cuda:0'), 'validation_loss': tensor(1.0497, device='cuda:0'), 'validation_loss_ce': tensor(0.5107, device='cuda:0'), 'validation_loss_bbox': tensor(0.0296, device='cuda:0'), 'validation_loss_giou': tensor(0.1956, device='cuda:0'), 'validation_cardinality_error': tensor(5.0769, device='cuda:0')}
{'training_loss': tensor(0.8323, device='cuda:0'), 'train_loss_ce': tensor(0.6898, device='cuda:0'), 'train_loss_bbox': tensor(0.0066, device='cuda:0'), 'train_loss_giou': tensor(0.0548, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(1.0601, device='cuda:0'), 'validation_loss_ce': tensor(0.5050, device='cuda:0'), 'validation_loss_bbox': tensor(0.0268, device='cuda:0'), 'validation_loss_giou': tensor(0.2104, device='cuda:0'), 'validation_cardinality_error': tensor(4.5385, device='cuda:0')}
{'training_loss': tensor(0.7580, device='cuda:0'), 'train_loss_ce': tensor(0.5615, device='cuda:0'), 'train_loss_bbox': tensor(0.0114, device='cuda:0'), 'train_loss_giou': tensor(0.0698, device='cuda:0'), 'train_cardinality_error': tensor(7., device='cuda:0'), 'validation_loss': tensor(0.9543, device='cuda:0'), 'validation_loss_ce': tensor(0.4833, device='cuda:0'), 'validation_loss_bbox': tensor(0.0251, device='cuda:0'), 'validation_loss_giou': tensor(0.1728, device='cuda:0'), 'validation_cardinality_error': tensor(3.6154, device='cuda:0')}
{'training_loss': tensor(0.3371, device='cuda:0'), 'train_loss_ce': tensor(0.3310, device='cuda:0'), 'train_loss_bbox': tensor(0.0007, device='cuda:0'), 'train_loss_giou': tensor(0.0012, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.1173, device='cuda:0'), 'validation_loss_ce': tensor(0.4800, device='cuda:0'), 'validation_loss_bbox': tensor(0.0358, device='cuda:0'), 'validation_loss_giou': tensor(0.2292, device='cuda:0'), 'validation_cardinality_error': tensor(3.0769, device='cuda:0')}
{'training_loss': tensor(0.5773, device='cuda:0'), 'train_loss_ce': tensor(0.4423, device='cuda:0'), 'train_loss_bbox': tensor(0.0132, device='cuda:0'), 'train_loss_giou': tensor(0.0345, device='cuda:0'), 'train_cardinality_error': tensor(2., device='cuda:0'), 'validation_loss': tensor(1.0636, device='cuda:0'), 'validation_loss_ce': tensor(0.4695, device='cuda:0'), 'validation_loss_bbox': tensor(0.0331, device='cuda:0'), 'validation_loss_giou': tensor(0.2142, device='cuda:0'), 'validation_cardinality_error': tensor(2.9231, device='cuda:0')}
{'training_loss': tensor(0.4857, device='cuda:0'), 'train_loss_ce': tensor(0.4557, device='cuda:0'), 'train_loss_bbox': tensor(0.0024, device='cuda:0'), 'train_loss_giou': tensor(0.0090, device='cuda:0'), 'train_cardinality_error': tensor(3., device='cuda:0'), 'validation_loss': tensor(0.9294, device='cuda:0'), 'validation_loss_ce': tensor(0.4525, device='cuda:0'), 'validation_loss_bbox': tensor(0.0261, device='cuda:0'), 'validation_loss_giou': tensor(0.1731, device='cuda:0'), 'validation_cardinality_error': tensor(2.6923, device='cuda:0')}
{'training_loss': tensor(0.2669, device='cuda:0'), 'train_loss_ce': tensor(0.2263, device='cuda:0'), 'train_loss_bbox': tensor(0.0032, device='cuda:0'), 'train_loss_giou': tensor(0.0124, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9155, device='cuda:0'), 'validation_loss_ce': tensor(0.4423, device='cuda:0'), 'validation_loss_bbox': tensor(0.0248, device='cuda:0'), 'validation_loss_giou': tensor(0.1747, device='cuda:0'), 'validation_cardinality_error': tensor(2.5385, device='cuda:0')}
{'training_loss': tensor(0.3390, device='cuda:0'), 'train_loss_ce': tensor(0.2219, device='cuda:0'), 'train_loss_bbox': tensor(0.0025, device='cuda:0'), 'train_loss_giou': tensor(0.0524, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9206, device='cuda:0'), 'validation_loss_ce': tensor(0.4341, device='cuda:0'), 'validation_loss_bbox': tensor(0.0263, device='cuda:0'), 'validation_loss_giou': tensor(0.1776, device='cuda:0'), 'validation_cardinality_error': tensor(2.3077, device='cuda:0')}
{'training_loss': tensor(0.2855, device='cuda:0'), 'train_loss_ce': tensor(0.2616, device='cuda:0'), 'train_loss_bbox': tensor(0.0020, device='cuda:0'), 'train_loss_giou': tensor(0.0070, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(1.0480, device='cuda:0'), 'validation_loss_ce': tensor(0.4292, device='cuda:0'), 'validation_loss_bbox': tensor(0.0343, device='cuda:0'), 'validation_loss_giou': tensor(0.2237, device='cuda:0'), 'validation_cardinality_error': tensor(2.2308, device='cuda:0')}
{'training_loss': tensor(0.2797, device='cuda:0'), 'train_loss_ce': tensor(0.2461, device='cuda:0'), 'train_loss_bbox': tensor(0.0029, device='cuda:0'), 'train_loss_giou': tensor(0.0096, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(1.0097, device='cuda:0'), 'validation_loss_ce': tensor(0.4160, device='cuda:0'), 'validation_loss_bbox': tensor(0.0317, device='cuda:0'), 'validation_loss_giou': tensor(0.2177, device='cuda:0'), 'validation_cardinality_error': tensor(2.0769, device='cuda:0')}
{'training_loss': tensor(0.4434, device='cuda:0'), 'train_loss_ce': tensor(0.3431, device='cuda:0'), 'train_loss_bbox': tensor(0.0047, device='cuda:0'), 'train_loss_giou': tensor(0.0385, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(0.9164, device='cuda:0'), 'validation_loss_ce': tensor(0.3956, device='cuda:0'), 'validation_loss_bbox': tensor(0.0263, device='cuda:0'), 'validation_loss_giou': tensor(0.1947, device='cuda:0'), 'validation_cardinality_error': tensor(1.6923, device='cuda:0')}
{'training_loss': tensor(0.7076, device='cuda:0'), 'train_loss_ce': tensor(0.5633, device='cuda:0'), 'train_loss_bbox': tensor(0.0063, device='cuda:0'), 'train_loss_giou': tensor(0.0564, device='cuda:0'), 'train_cardinality_error': tensor(5., device='cuda:0'), 'validation_loss': tensor(1.0224, device='cuda:0'), 'validation_loss_ce': tensor(0.3994, device='cuda:0'), 'validation_loss_bbox': tensor(0.0360, device='cuda:0'), 'validation_loss_giou': tensor(0.2216, device='cuda:0'), 'validation_cardinality_error': tensor(1.8462, device='cuda:0')}
{'training_loss': tensor(0.3618, device='cuda:0'), 'train_loss_ce': tensor(0.2966, device='cuda:0'), 'train_loss_bbox': tensor(0.0034, device='cuda:0'), 'train_loss_giou': tensor(0.0242, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(0.9028, device='cuda:0'), 'validation_loss_ce': tensor(0.3809, device='cuda:0'), 'validation_loss_bbox': tensor(0.0293, device='cuda:0'), 'validation_loss_giou': tensor(0.1877, device='cuda:0'), 'validation_cardinality_error': tensor(1.8462, device='cuda:0')}
{'training_loss': tensor(0.5109, device='cuda:0'), 'train_loss_ce': tensor(0.3869, device='cuda:0'), 'train_loss_bbox': tensor(0.0057, device='cuda:0'), 'train_loss_giou': tensor(0.0478, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(0.9129, device='cuda:0'), 'validation_loss_ce': tensor(0.3762, device='cuda:0'), 'validation_loss_bbox': tensor(0.0264, device='cuda:0'), 'validation_loss_giou': tensor(0.2024, device='cuda:0'), 'validation_cardinality_error': tensor(1.6923, device='cuda:0')}
{'training_loss': tensor(0.3791, device='cuda:0'), 'train_loss_ce': tensor(0.3609, device='cuda:0'), 'train_loss_bbox': tensor(0.0016, device='cuda:0'), 'train_loss_giou': tensor(0.0050, device='cuda:0'), 'train_cardinality_error': tensor(0., device='cuda:0'), 'validation_loss': tensor(0.9797, device='cuda:0'), 'validation_loss_ce': tensor(0.3611, device='cuda:0'), 'validation_loss_bbox': tensor(0.0331, device='cuda:0'), 'validation_loss_giou': tensor(0.2265, device='cuda:0'), 'validation_cardinality_error': tensor(1.6923, device='cuda:0')}
{'training_loss': tensor(0.2169, device='cuda:0'), 'train_loss_ce': tensor(0.1839, device='cuda:0'), 'train_loss_bbox': tensor(0.0030, device='cuda:0'), 'train_loss_giou': tensor(0.0091, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8334, device='cuda:0'), 'validation_loss_ce': tensor(0.3444, device='cuda:0'), 'validation_loss_bbox': tensor(0.0261, device='cuda:0'), 'validation_loss_giou': tensor(0.1792, device='cuda:0'), 'validation_cardinality_error': tensor(1.7692, device='cuda:0')}

Examples

{'size': tensor([ 800, 1066]), 'image_id': tensor([0]), 'class_labels': tensor([0]), 'boxes': tensor([[0.5955, 0.5811, 0.2202, 0.3561]]), 'area': tensor([3681.5083]), 'iscrowd': tensor([0]), 'orig_size': tensor([1536, 2048])}

MedicalVision/detr_balloon_example