---
library_name: transformers
tags: []
---

## Original result
```
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.007
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.009
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.009
```

## After training result
```
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.058
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.108
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.054
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.019
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.051
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.099
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.170
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.192
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.046
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.177
```

## Config
- dataset: NIH
- original model: hustvl/yolos-tiny
- lr: 0.0001
- dropout_rate: 0.1
- weight_decay: 0.001
- max_epochs: 100
- train samples: 885

## Logging
### Training process
```
{'validation_loss': tensor(7.2045, device='cuda:0'), 'validation_loss_ce': tensor(2.2968, device='cuda:0'), 'validation_loss_bbox': tensor(0.5672, device='cuda:0'), 'validation_loss_giou': tensor(1.0358, device='cuda:0'), 'validation_cardinality_error': tensor(98.9688, device='cuda:0')}
{'training_loss': tensor(2.5200, device='cuda:0'), 'train_loss_ce': tensor(0.4957, device='cuda:0'), 'train_loss_bbox': tensor(0.1564, device='cuda:0'), 'train_loss_giou': tensor(0.6212, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2375, device='cuda:0'), 'validation_loss_ce': tensor(0.4545, device='cuda:0'), 'validation_loss_bbox': tensor(0.1507, device='cuda:0'), 'validation_loss_giou': tensor(0.5146, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(2.2501, device='cuda:0'), 'train_loss_ce': tensor(0.4415, device='cuda:0'), 'train_loss_bbox': tensor(0.1535, device='cuda:0'), 'train_loss_giou': tensor(0.5207, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2336, device='cuda:0'), 'validation_loss_ce': tensor(0.4277, device='cuda:0'), 'validation_loss_bbox': tensor(0.1595, device='cuda:0'), 'validation_loss_giou': tensor(0.5043, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(2.4403, device='cuda:0'), 'train_loss_ce': tensor(0.3938, device='cuda:0'), 'train_loss_bbox': tensor(0.1813, device='cuda:0'), 'train_loss_giou': tensor(0.5700, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7168, device='cuda:0'), 'validation_loss_ce': tensor(0.4196, device='cuda:0'), 'validation_loss_bbox': tensor(0.2084, device='cuda:0'), 'validation_loss_giou': tensor(0.6275, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(2.2682, device='cuda:0'), 'train_loss_ce': tensor(0.4668, device='cuda:0'), 'train_loss_bbox': tensor(0.1324, device='cuda:0'), 'train_loss_giou': tensor(0.5698, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.3310, device='cuda:0'), 'validation_loss_ce': tensor(0.4273, device='cuda:0'), 'validation_loss_bbox': tensor(0.1675, device='cuda:0'), 'validation_loss_giou': tensor(0.5331, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(2.1533, device='cuda:0'), 'train_loss_ce': tensor(0.3045, device='cuda:0'), 'train_loss_bbox': tensor(0.1709, device='cuda:0'), 'train_loss_giou': tensor(0.4971, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.1458, device='cuda:0'), 'validation_loss_ce': tensor(0.4185, device='cuda:0'), 'validation_loss_bbox': tensor(0.1446, device='cuda:0'), 'validation_loss_giou': tensor(0.5023, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(1.6247, device='cuda:0'), 'train_loss_ce': tensor(0.2961, device='cuda:0'), 'train_loss_bbox': tensor(0.0975, device='cuda:0'), 'train_loss_giou': tensor(0.4206, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.1055, device='cuda:0'), 'validation_loss_ce': tensor(0.4085, device='cuda:0'), 'validation_loss_bbox': tensor(0.1438, device='cuda:0'), 'validation_loss_giou': tensor(0.4889, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(1.5941, device='cuda:0'), 'train_loss_ce': tensor(0.3669, device='cuda:0'), 'train_loss_bbox': tensor(0.0878, device='cuda:0'), 'train_loss_giou': tensor(0.3941, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0741, device='cuda:0'), 'validation_loss_ce': tensor(0.4017, device='cuda:0'), 'validation_loss_bbox': tensor(0.1364, device='cuda:0'), 'validation_loss_giou': tensor(0.4951, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(1.7629, device='cuda:0'), 'train_loss_ce': tensor(0.3858, device='cuda:0'), 'train_loss_bbox': tensor(0.1226, device='cuda:0'), 'train_loss_giou': tensor(0.3820, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0028, device='cuda:0'), 'validation_loss_ce': tensor(0.3956, device='cuda:0'), 'validation_loss_bbox': tensor(0.1363, device='cuda:0'), 'validation_loss_giou': tensor(0.4629, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(1.6581, device='cuda:0'), 'train_loss_ce': tensor(0.3119, device='cuda:0'), 'train_loss_bbox': tensor(0.0921, device='cuda:0'), 'train_loss_giou': tensor(0.4429, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9400, device='cuda:0'), 'validation_loss_ce': tensor(0.3832, device='cuda:0'), 'validation_loss_bbox': tensor(0.1252, device='cuda:0'), 'validation_loss_giou': tensor(0.4655, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(1.3849, device='cuda:0'), 'train_loss_ce': tensor(0.3898, device='cuda:0'), 'train_loss_bbox': tensor(0.0691, device='cuda:0'), 'train_loss_giou': tensor(0.3247, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8570, device='cuda:0'), 'validation_loss_ce': tensor(0.3680, device='cuda:0'), 'validation_loss_bbox': tensor(0.1198, device='cuda:0'), 'validation_loss_giou': tensor(0.4451, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(2.3841, device='cuda:0'), 'train_loss_ce': tensor(0.4545, device='cuda:0'), 'train_loss_bbox': tensor(0.1736, device='cuda:0'), 'train_loss_giou': tensor(0.5307, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9082, device='cuda:0'), 'validation_loss_ce': tensor(0.3641, device='cuda:0'), 'validation_loss_bbox': tensor(0.1256, device='cuda:0'), 'validation_loss_giou': tensor(0.4582, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(1.6018, device='cuda:0'), 'train_loss_ce': tensor(0.3856, device='cuda:0'), 'train_loss_bbox': tensor(0.0900, device='cuda:0'), 'train_loss_giou': tensor(0.3831, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8398, device='cuda:0'), 'validation_loss_ce': tensor(0.3599, device='cuda:0'), 'validation_loss_bbox': tensor(0.1200, device='cuda:0'), 'validation_loss_giou': tensor(0.4398, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
{'training_loss': tensor(2.1644, device='cuda:0'), 'train_loss_ce': tensor(0.3639, device='cuda:0'), 'train_loss_bbox': tensor(0.1234, device='cuda:0'), 'train_loss_giou': tensor(0.5917, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8483, device='cuda:0'), 'validation_loss_ce': tensor(0.3541, device='cuda:0'), 'validation_loss_bbox': tensor(0.1239, device='cuda:0'), 'validation_loss_giou': tensor(0.4373, device='cuda:0'), 'validation_cardinality_error': tensor(0.7980, device='cuda:0')}
{'training_loss': tensor(1.4989, device='cuda:0'), 'train_loss_ce': tensor(0.3999, device='cuda:0'), 'train_loss_bbox': tensor(0.0872, device='cuda:0'), 'train_loss_giou': tensor(0.3315, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.8334, device='cuda:0'), 'validation_loss_ce': tensor(0.3644, device='cuda:0'), 'validation_loss_bbox': tensor(0.1196, device='cuda:0'), 'validation_loss_giou': tensor(0.4356, device='cuda:0'), 'validation_cardinality_error': tensor(0.5354, device='cuda:0')}
{'training_loss': tensor(1.2876, device='cuda:0'), 'train_loss_ce': tensor(0.3680, device='cuda:0'), 'train_loss_bbox': tensor(0.0884, device='cuda:0'), 'train_loss_giou': tensor(0.2388, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8002, device='cuda:0'), 'validation_loss_ce': tensor(0.3446, device='cuda:0'), 'validation_loss_bbox': tensor(0.1209, device='cuda:0'), 'validation_loss_giou': tensor(0.4255, device='cuda:0'), 'validation_cardinality_error': tensor(0.7273, device='cuda:0')}
{'training_loss': tensor(1.8735, device='cuda:0'), 'train_loss_ce': tensor(0.3472, device='cuda:0'), 'train_loss_bbox': tensor(0.1311, device='cuda:0'), 'train_loss_giou': tensor(0.4354, device='cuda:0'), 'train_cardinality_error': tensor(0.2000, device='cuda:0'), 'validation_loss': tensor(2.0753, device='cuda:0'), 'validation_loss_ce': tensor(0.3446, device='cuda:0'), 'validation_loss_bbox': tensor(0.1430, device='cuda:0'), 'validation_loss_giou': tensor(0.5078, device='cuda:0'), 'validation_cardinality_error': tensor(0.4747, device='cuda:0')}
{'training_loss': tensor(1.9574, device='cuda:0'), 'train_loss_ce': tensor(0.3733, device='cuda:0'), 'train_loss_bbox': tensor(0.1423, device='cuda:0'), 'train_loss_giou': tensor(0.4363, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9587, device='cuda:0'), 'validation_loss_ce': tensor(0.3564, device='cuda:0'), 'validation_loss_bbox': tensor(0.1323, device='cuda:0'), 'validation_loss_giou': tensor(0.4704, device='cuda:0'), 'validation_cardinality_error': tensor(0.8788, device='cuda:0')}
{'training_loss': tensor(1.4950, device='cuda:0'), 'train_loss_ce': tensor(0.2871, device='cuda:0'), 'train_loss_bbox': tensor(0.0836, device='cuda:0'), 'train_loss_giou': tensor(0.3950, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9386, device='cuda:0'), 'validation_loss_ce': tensor(0.3506, device='cuda:0'), 'validation_loss_bbox': tensor(0.1328, device='cuda:0'), 'validation_loss_giou': tensor(0.4620, device='cuda:0'), 'validation_cardinality_error': tensor(0.7576, device='cuda:0')}
{'training_loss': tensor(1.1535, device='cuda:0'), 'train_loss_ce': tensor(0.2828, device='cuda:0'), 'train_loss_bbox': tensor(0.0519, device='cuda:0'), 'train_loss_giou': tensor(0.3056, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8730, device='cuda:0'), 'validation_loss_ce': tensor(0.3456, device='cuda:0'), 'validation_loss_bbox': tensor(0.1236, device='cuda:0'), 'validation_loss_giou': tensor(0.4546, device='cuda:0'), 'validation_cardinality_error': tensor(0.6162, device='cuda:0')}
{'training_loss': tensor(1.4150, device='cuda:0'), 'train_loss_ce': tensor(0.1594, device='cuda:0'), 'train_loss_bbox': tensor(0.1283, device='cuda:0'), 'train_loss_giou': tensor(0.3072, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8677, device='cuda:0'), 'validation_loss_ce': tensor(0.3425, device='cuda:0'), 'validation_loss_bbox': tensor(0.1262, device='cuda:0'), 'validation_loss_giou': tensor(0.4470, device='cuda:0'), 'validation_cardinality_error': tensor(0.5758, device='cuda:0')}
{'training_loss': tensor(1.1482, device='cuda:0'), 'train_loss_ce': tensor(0.2020, device='cuda:0'), 'train_loss_bbox': tensor(0.0582, device='cuda:0'), 'train_loss_giou': tensor(0.3276, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8096, device='cuda:0'), 'validation_loss_ce': tensor(0.3357, device='cuda:0'), 'validation_loss_bbox': tensor(0.1197, device='cuda:0'), 'validation_loss_giou': tensor(0.4378, device='cuda:0'), 'validation_cardinality_error': tensor(0.5253, device='cuda:0')}
{'training_loss': tensor(1.5939, device='cuda:0'), 'train_loss_ce': tensor(0.3802, device='cuda:0'), 'train_loss_bbox': tensor(0.0916, device='cuda:0'), 'train_loss_giou': tensor(0.3778, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.9104, device='cuda:0'), 'validation_loss_ce': tensor(0.3441, device='cuda:0'), 'validation_loss_bbox': tensor(0.1264, device='cuda:0'), 'validation_loss_giou': tensor(0.4671, device='cuda:0'), 'validation_cardinality_error': tensor(0.4343, device='cuda:0')}
{'training_loss': tensor(1.2458, device='cuda:0'), 'train_loss_ce': tensor(0.3179, device='cuda:0'), 'train_loss_bbox': tensor(0.0807, device='cuda:0'), 'train_loss_giou': tensor(0.2623, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.7529, device='cuda:0'), 'validation_loss_ce': tensor(0.3379, device='cuda:0'), 'validation_loss_bbox': tensor(0.1140, device='cuda:0'), 'validation_loss_giou': tensor(0.4224, device='cuda:0'), 'validation_cardinality_error': tensor(0.5354, device='cuda:0')}
{'training_loss': tensor(1.2705, device='cuda:0'), 'train_loss_ce': tensor(0.2641, device='cuda:0'), 'train_loss_bbox': tensor(0.0617, device='cuda:0'), 'train_loss_giou': tensor(0.3488, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8156, device='cuda:0'), 'validation_loss_ce': tensor(0.3391, device='cuda:0'), 'validation_loss_bbox': tensor(0.1156, device='cuda:0'), 'validation_loss_giou': tensor(0.4492, device='cuda:0'), 'validation_cardinality_error': tensor(0.7374, device='cuda:0')}
{'training_loss': tensor(1.3914, device='cuda:0'), 'train_loss_ce': tensor(0.2143, device='cuda:0'), 'train_loss_bbox': tensor(0.0970, device='cuda:0'), 'train_loss_giou': tensor(0.3462, device='cuda:0'), 'train_cardinality_error': tensor(0.2000, device='cuda:0'), 'validation_loss': tensor(1.7497, device='cuda:0'), 'validation_loss_ce': tensor(0.3505, device='cuda:0'), 'validation_loss_bbox': tensor(0.1131, device='cuda:0'), 'validation_loss_giou': tensor(0.4170, device='cuda:0'), 'validation_cardinality_error': tensor(0.5960, device='cuda:0')}
{'training_loss': tensor(1.6075, device='cuda:0'), 'train_loss_ce': tensor(0.3839, device='cuda:0'), 'train_loss_bbox': tensor(0.0690, device='cuda:0'), 'train_loss_giou': tensor(0.4394, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.6896, device='cuda:0'), 'validation_loss_ce': tensor(0.3293, device='cuda:0'), 'validation_loss_bbox': tensor(0.1059, device='cuda:0'), 'validation_loss_giou': tensor(0.4154, device='cuda:0'), 'validation_cardinality_error': tensor(0.3636, device='cuda:0')}
{'training_loss': tensor(1.7218, device='cuda:0'), 'train_loss_ce': tensor(0.3454, device='cuda:0'), 'train_loss_bbox': tensor(0.1221, device='cuda:0'), 'train_loss_giou': tensor(0.3830, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.7517, device='cuda:0'), 'validation_loss_ce': tensor(0.3381, device='cuda:0'), 'validation_loss_bbox': tensor(0.1116, device='cuda:0'), 'validation_loss_giou': tensor(0.4279, device='cuda:0'), 'validation_cardinality_error': tensor(0.5859, device='cuda:0')}
{'training_loss': tensor(1.4428, device='cuda:0'), 'train_loss_ce': tensor(0.2685, device='cuda:0'), 'train_loss_bbox': tensor(0.0879, device='cuda:0'), 'train_loss_giou': tensor(0.3674, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.7905, device='cuda:0'), 'validation_loss_ce': tensor(0.3374, device='cuda:0'), 'validation_loss_bbox': tensor(0.1187, device='cuda:0'), 'validation_loss_giou': tensor(0.4298, device='cuda:0'), 'validation_cardinality_error': tensor(0.6162, device='cuda:0')}
{'training_loss': tensor(1.3059, device='cuda:0'), 'train_loss_ce': tensor(0.3185, device='cuda:0'), 'train_loss_bbox': tensor(0.0537, device='cuda:0'), 'train_loss_giou': tensor(0.3596, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.7305, device='cuda:0'), 'validation_loss_ce': tensor(0.3429, device='cuda:0'), 'validation_loss_bbox': tensor(0.1116, device='cuda:0'), 'validation_loss_giou': tensor(0.4149, device='cuda:0'), 'validation_cardinality_error': tensor(0.6364, device='cuda:0')}
{'training_loss': tensor(1.4403, device='cuda:0'), 'train_loss_ce': tensor(0.3263, device='cuda:0'), 'train_loss_bbox': tensor(0.1015, device='cuda:0'), 'train_loss_giou': tensor(0.3034, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.9518, device='cuda:0'), 'validation_loss_ce': tensor(0.3443, device='cuda:0'), 'validation_loss_bbox': tensor(0.1318, device='cuda:0'), 'validation_loss_giou': tensor(0.4743, device='cuda:0'), 'validation_cardinality_error': tensor(0.7172, device='cuda:0')}
{'training_loss': tensor(1.1981, device='cuda:0'), 'train_loss_ce': tensor(0.2236, device='cuda:0'), 'train_loss_bbox': tensor(0.0807, device='cuda:0'), 'train_loss_giou': tensor(0.2854, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8668, device='cuda:0'), 'validation_loss_ce': tensor(0.3564, device='cuda:0'), 'validation_loss_bbox': tensor(0.1230, device='cuda:0'), 'validation_loss_giou': tensor(0.4476, device='cuda:0'), 'validation_cardinality_error': tensor(0.7980, device='cuda:0')}
{'training_loss': tensor(1.5449, device='cuda:0'), 'train_loss_ce': tensor(0.2428, device='cuda:0'), 'train_loss_bbox': tensor(0.0945, device='cuda:0'), 'train_loss_giou': tensor(0.4148, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8501, device='cuda:0'), 'validation_loss_ce': tensor(0.3724, device='cuda:0'), 'validation_loss_bbox': tensor(0.1165, device='cuda:0'), 'validation_loss_giou': tensor(0.4475, device='cuda:0'), 'validation_cardinality_error': tensor(0.5758, device='cuda:0')}
{'training_loss': tensor(1.7828, device='cuda:0'), 'train_loss_ce': tensor(0.4551, device='cuda:0'), 'train_loss_bbox': tensor(0.1038, device='cuda:0'), 'train_loss_giou': tensor(0.4043, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8635, device='cuda:0'), 'validation_loss_ce': tensor(0.3396, device='cuda:0'), 'validation_loss_bbox': tensor(0.1237, device='cuda:0'), 'validation_loss_giou': tensor(0.4528, device='cuda:0'), 'validation_cardinality_error': tensor(0.6566, device='cuda:0')}
{'training_loss': tensor(1.1647, device='cuda:0'), 'train_loss_ce': tensor(0.2744, device='cuda:0'), 'train_loss_bbox': tensor(0.0722, device='cuda:0'), 'train_loss_giou': tensor(0.2647, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.7792, device='cuda:0'), 'validation_loss_ce': tensor(0.3315, device='cuda:0'), 'validation_loss_bbox': tensor(0.1158, device='cuda:0'), 'validation_loss_giou': tensor(0.4344, device='cuda:0'), 'validation_cardinality_error': tensor(0.7980, device='cuda:0')}
{'training_loss': tensor(1.1196, device='cuda:0'), 'train_loss_ce': tensor(0.1983, device='cuda:0'), 'train_loss_bbox': tensor(0.0853, device='cuda:0'), 'train_loss_giou': tensor(0.2474, device='cuda:0'), 'train_cardinality_error': tensor(1.2000, device='cuda:0'), 'validation_loss': tensor(1.7475, device='cuda:0'), 'validation_loss_ce': tensor(0.3421, device='cuda:0'), 'validation_loss_bbox': tensor(0.1131, device='cuda:0'), 'validation_loss_giou': tensor(0.4200, device='cuda:0'), 'validation_cardinality_error': tensor(0.6768, device='cuda:0')}
{'training_loss': tensor(1.1440, device='cuda:0'), 'train_loss_ce': tensor(0.1990, device='cuda:0'), 'train_loss_bbox': tensor(0.0908, device='cuda:0'), 'train_loss_giou': tensor(0.2456, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.7808, device='cuda:0'), 'validation_loss_ce': tensor(0.3481, device='cuda:0'), 'validation_loss_bbox': tensor(0.1152, device='cuda:0'), 'validation_loss_giou': tensor(0.4284, device='cuda:0'), 'validation_cardinality_error': tensor(0.7071, device='cuda:0')}
{'training_loss': tensor(1.2752, device='cuda:0'), 'train_loss_ce': tensor(0.1843, device='cuda:0'), 'train_loss_bbox': tensor(0.1018, device='cuda:0'), 'train_loss_giou': tensor(0.2909, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.8421, device='cuda:0'), 'validation_loss_ce': tensor(0.3465, device='cuda:0'), 'validation_loss_bbox': tensor(0.1200, device='cuda:0'), 'validation_loss_giou': tensor(0.4478, device='cuda:0'), 'validation_cardinality_error': tensor(0.5556, device='cuda:0')}
{'training_loss': tensor(1.4315, device='cuda:0'), 'train_loss_ce': tensor(0.2973, device='cuda:0'), 'train_loss_bbox': tensor(0.0943, device='cuda:0'), 'train_loss_giou': tensor(0.3314, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.7926, device='cuda:0'), 'validation_loss_ce': tensor(0.3650, device='cuda:0'), 'validation_loss_bbox': tensor(0.1166, device='cuda:0'), 'validation_loss_giou': tensor(0.4223, device='cuda:0'), 'validation_cardinality_error': tensor(0.5455, device='cuda:0')}
{'training_loss': tensor(1.1553, device='cuda:0'), 'train_loss_ce': tensor(0.2358, device='cuda:0'), 'train_loss_bbox': tensor(0.0676, device='cuda:0'), 'train_loss_giou': tensor(0.2907, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8522, device='cuda:0'), 'validation_loss_ce': tensor(0.3531, device='cuda:0'), 'validation_loss_bbox': tensor(0.1244, device='cuda:0'), 'validation_loss_giou': tensor(0.4386, device='cuda:0'), 'validation_cardinality_error': tensor(0.6869, device='cuda:0')}
{'training_loss': tensor(1.1894, device='cuda:0'), 'train_loss_ce': tensor(0.2374, device='cuda:0'), 'train_loss_bbox': tensor(0.0772, device='cuda:0'), 'train_loss_giou': tensor(0.2831, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8030, device='cuda:0'), 'validation_loss_ce': tensor(0.3644, device='cuda:0'), 'validation_loss_bbox': tensor(0.1184, device='cuda:0'), 'validation_loss_giou': tensor(0.4234, device='cuda:0'), 'validation_cardinality_error': tensor(0.6970, device='cuda:0')}
{'training_loss': tensor(1.1038, device='cuda:0'), 'train_loss_ce': tensor(0.2761, device='cuda:0'), 'train_loss_bbox': tensor(0.0691, device='cuda:0'), 'train_loss_giou': tensor(0.2410, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8602, device='cuda:0'), 'validation_loss_ce': tensor(0.3383, device='cuda:0'), 'validation_loss_bbox': tensor(0.1233, device='cuda:0'), 'validation_loss_giou': tensor(0.4526, device='cuda:0'), 'validation_cardinality_error': tensor(0.5657, device='cuda:0')}
{'training_loss': tensor(1.6817, device='cuda:0'), 'train_loss_ce': tensor(0.3073, device='cuda:0'), 'train_loss_bbox': tensor(0.1129, device='cuda:0'), 'train_loss_giou': tensor(0.4049, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9009, device='cuda:0'), 'validation_loss_ce': tensor(0.3715, device='cuda:0'), 'validation_loss_bbox': tensor(0.1227, device='cuda:0'), 'validation_loss_giou': tensor(0.4580, device='cuda:0'), 'validation_cardinality_error': tensor(0.6162, device='cuda:0')}
{'training_loss': tensor(1.3429, device='cuda:0'), 'train_loss_ce': tensor(0.2732, device='cuda:0'), 'train_loss_bbox': tensor(0.0755, device='cuda:0'), 'train_loss_giou': tensor(0.3461, device='cuda:0'), 'train_cardinality_error': tensor(0.2000, device='cuda:0'), 'validation_loss': tensor(1.7967, device='cuda:0'), 'validation_loss_ce': tensor(0.3434, device='cuda:0'), 'validation_loss_bbox': tensor(0.1215, device='cuda:0'), 'validation_loss_giou': tensor(0.4229, device='cuda:0'), 'validation_cardinality_error': tensor(0.6061, device='cuda:0')}
{'training_loss': tensor(1.1241, device='cuda:0'), 'train_loss_ce': tensor(0.2168, device='cuda:0'), 'train_loss_bbox': tensor(0.0621, device='cuda:0'), 'train_loss_giou': tensor(0.2985, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.7612, device='cuda:0'), 'validation_loss_ce': tensor(0.3442, device='cuda:0'), 'validation_loss_bbox': tensor(0.1143, device='cuda:0'), 'validation_loss_giou': tensor(0.4227, device='cuda:0'), 'validation_cardinality_error': tensor(0.6667, device='cuda:0')}
{'training_loss': tensor(0.9909, device='cuda:0'), 'train_loss_ce': tensor(0.3210, device='cuda:0'), 'train_loss_bbox': tensor(0.0531, device='cuda:0'), 'train_loss_giou': tensor(0.2022, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8190, device='cuda:0'), 'validation_loss_ce': tensor(0.3695, device='cuda:0'), 'validation_loss_bbox': tensor(0.1170, device='cuda:0'), 'validation_loss_giou': tensor(0.4323, device='cuda:0'), 'validation_cardinality_error': tensor(0.6566, device='cuda:0')}
{'training_loss': tensor(1.7008, device='cuda:0'), 'train_loss_ce': tensor(0.2614, device='cuda:0'), 'train_loss_bbox': tensor(0.1000, device='cuda:0'), 'train_loss_giou': tensor(0.4698, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9067, device='cuda:0'), 'validation_loss_ce': tensor(0.3768, device='cuda:0'), 'validation_loss_bbox': tensor(0.1280, device='cuda:0'), 'validation_loss_giou': tensor(0.4449, device='cuda:0'), 'validation_cardinality_error': tensor(0.5556, device='cuda:0')}
{'training_loss': tensor(1.1680, device='cuda:0'), 'train_loss_ce': tensor(0.3623, device='cuda:0'), 'train_loss_bbox': tensor(0.0808, device='cuda:0'), 'train_loss_giou': tensor(0.2009, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8161, device='cuda:0'), 'validation_loss_ce': tensor(0.3561, device='cuda:0'), 'validation_loss_bbox': tensor(0.1127, device='cuda:0'), 'validation_loss_giou': tensor(0.4482, device='cuda:0'), 'validation_cardinality_error': tensor(0.6364, device='cuda:0')}
{'training_loss': tensor(1.3337, device='cuda:0'), 'train_loss_ce': tensor(0.2750, device='cuda:0'), 'train_loss_bbox': tensor(0.0752, device='cuda:0'), 'train_loss_giou': tensor(0.3413, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9931, device='cuda:0'), 'validation_loss_ce': tensor(0.3794, device='cuda:0'), 'validation_loss_bbox': tensor(0.1348, device='cuda:0'), 'validation_loss_giou': tensor(0.4698, device='cuda:0'), 'validation_cardinality_error': tensor(0.7273, device='cuda:0')}
{'training_loss': tensor(1.6290, device='cuda:0'), 'train_loss_ce': tensor(0.1759, device='cuda:0'), 'train_loss_bbox': tensor(0.1032, device='cuda:0'), 'train_loss_giou': tensor(0.4685, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8118, device='cuda:0'), 'validation_loss_ce': tensor(0.3543, device='cuda:0'), 'validation_loss_bbox': tensor(0.1188, device='cuda:0'), 'validation_loss_giou': tensor(0.4318, device='cuda:0'), 'validation_cardinality_error': tensor(0.8182, device='cuda:0')}
{'training_loss': tensor(1.0428, device='cuda:0'), 'train_loss_ce': tensor(0.2748, device='cuda:0'), 'train_loss_bbox': tensor(0.0603, device='cuda:0'), 'train_loss_giou': tensor(0.2332, device='cuda:0'), 'train_cardinality_error': tensor(0.2000, device='cuda:0'), 'validation_loss': tensor(1.8947, device='cuda:0'), 'validation_loss_ce': tensor(0.3599, device='cuda:0'), 'validation_loss_bbox': tensor(0.1270, device='cuda:0'), 'validation_loss_giou': tensor(0.4498, device='cuda:0'), 'validation_cardinality_error': tensor(0.7071, device='cuda:0')}
{'training_loss': tensor(1.3277, device='cuda:0'), 'train_loss_ce': tensor(0.3765, device='cuda:0'), 'train_loss_bbox': tensor(0.0675, device='cuda:0'), 'train_loss_giou': tensor(0.3067, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8379, device='cuda:0'), 'validation_loss_ce': tensor(0.3622, device='cuda:0'), 'validation_loss_bbox': tensor(0.1199, device='cuda:0'), 'validation_loss_giou': tensor(0.4380, device='cuda:0'), 'validation_cardinality_error': tensor(0.6768, device='cuda:0')}
{'training_loss': tensor(0.8329, device='cuda:0'), 'train_loss_ce': tensor(0.1931, device='cuda:0'), 'train_loss_bbox': tensor(0.0430, device='cuda:0'), 'train_loss_giou': tensor(0.2123, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8473, device='cuda:0'), 'validation_loss_ce': tensor(0.3623, device='cuda:0'), 'validation_loss_bbox': tensor(0.1220, device='cuda:0'), 'validation_loss_giou': tensor(0.4375, device='cuda:0'), 'validation_cardinality_error': tensor(0.6970, device='cuda:0')}
{'training_loss': tensor(0.8308, device='cuda:0'), 'train_loss_ce': tensor(0.2576, device='cuda:0'), 'train_loss_bbox': tensor(0.0404, device='cuda:0'), 'train_loss_giou': tensor(0.1856, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8586, device='cuda:0'), 'validation_loss_ce': tensor(0.3569, device='cuda:0'), 'validation_loss_bbox': tensor(0.1232, device='cuda:0'), 'validation_loss_giou': tensor(0.4428, device='cuda:0'), 'validation_cardinality_error': tensor(0.8182, device='cuda:0')}
{'training_loss': tensor(1.0060, device='cuda:0'), 'train_loss_ce': tensor(0.2221, device='cuda:0'), 'train_loss_bbox': tensor(0.0567, device='cuda:0'), 'train_loss_giou': tensor(0.2502, device='cuda:0'), 'train_cardinality_error': tensor(1.8000, device='cuda:0'), 'validation_loss': tensor(1.8269, device='cuda:0'), 'validation_loss_ce': tensor(0.3623, device='cuda:0'), 'validation_loss_bbox': tensor(0.1209, device='cuda:0'), 'validation_loss_giou': tensor(0.4299, device='cuda:0'), 'validation_cardinality_error': tensor(0.6566, device='cuda:0')}
{'training_loss': tensor(0.8133, device='cuda:0'), 'train_loss_ce': tensor(0.1912, device='cuda:0'), 'train_loss_bbox': tensor(0.0634, device='cuda:0'), 'train_loss_giou': tensor(0.1526, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.8257, device='cuda:0'), 'validation_loss_ce': tensor(0.3930, device='cuda:0'), 'validation_loss_bbox': tensor(0.1178, device='cuda:0'), 'validation_loss_giou': tensor(0.4218, device='cuda:0'), 'validation_cardinality_error': tensor(0.7475, device='cuda:0')}
{'training_loss': tensor(0.7469, device='cuda:0'), 'train_loss_ce': tensor(0.1062, device='cuda:0'), 'train_loss_bbox': tensor(0.0483, device='cuda:0'), 'train_loss_giou': tensor(0.1995, device='cuda:0'), 'train_cardinality_error': tensor(1.6000, device='cuda:0'), 'validation_loss': tensor(1.8782, device='cuda:0'), 'validation_loss_ce': tensor(0.3785, device='cuda:0'), 'validation_loss_bbox': tensor(0.1186, device='cuda:0'), 'validation_loss_giou': tensor(0.4534, device='cuda:0'), 'validation_cardinality_error': tensor(0.7172, device='cuda:0')}
{'training_loss': tensor(0.9501, device='cuda:0'), 'train_loss_ce': tensor(0.1827, device='cuda:0'), 'train_loss_bbox': tensor(0.0738, device='cuda:0'), 'train_loss_giou': tensor(0.1992, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9531, device='cuda:0'), 'validation_loss_ce': tensor(0.3779, device='cuda:0'), 'validation_loss_bbox': tensor(0.1274, device='cuda:0'), 'validation_loss_giou': tensor(0.4690, device='cuda:0'), 'validation_cardinality_error': tensor(0.7677, device='cuda:0')}
{'training_loss': tensor(0.6456, device='cuda:0'), 'train_loss_ce': tensor(0.1112, device='cuda:0'), 'train_loss_bbox': tensor(0.0401, device='cuda:0'), 'train_loss_giou': tensor(0.1668, device='cuda:0'), 'train_cardinality_error': tensor(0.2000, device='cuda:0'), 'validation_loss': tensor(1.9024, device='cuda:0'), 'validation_loss_ce': tensor(0.3739, device='cuda:0'), 'validation_loss_bbox': tensor(0.1244, device='cuda:0'), 'validation_loss_giou': tensor(0.4531, device='cuda:0'), 'validation_cardinality_error': tensor(0.7071, device='cuda:0')}
{'training_loss': tensor(0.7231, device='cuda:0'), 'train_loss_ce': tensor(0.2216, device='cuda:0'), 'train_loss_bbox': tensor(0.0425, device='cuda:0'), 'train_loss_giou': tensor(0.1446, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8880, device='cuda:0'), 'validation_loss_ce': tensor(0.3662, device='cuda:0'), 'validation_loss_bbox': tensor(0.1263, device='cuda:0'), 'validation_loss_giou': tensor(0.4451, device='cuda:0'), 'validation_cardinality_error': tensor(0.6667, device='cuda:0')}
{'training_loss': tensor(0.8083, device='cuda:0'), 'train_loss_ce': tensor(0.1597, device='cuda:0'), 'train_loss_bbox': tensor(0.0396, device='cuda:0'), 'train_loss_giou': tensor(0.2252, device='cuda:0'), 'train_cardinality_error': tensor(1.2000, device='cuda:0'), 'validation_loss': tensor(1.8730, device='cuda:0'), 'validation_loss_ce': tensor(0.3666, device='cuda:0'), 'validation_loss_bbox': tensor(0.1243, device='cuda:0'), 'validation_loss_giou': tensor(0.4424, device='cuda:0'), 'validation_cardinality_error': tensor(0.7677, device='cuda:0')}
{'training_loss': tensor(0.8300, device='cuda:0'), 'train_loss_ce': tensor(0.1294, device='cuda:0'), 'train_loss_bbox': tensor(0.0581, device='cuda:0'), 'train_loss_giou': tensor(0.2051, device='cuda:0'), 'train_cardinality_error': tensor(2.4000, device='cuda:0'), 'validation_loss': tensor(1.8896, device='cuda:0'), 'validation_loss_ce': tensor(0.3598, device='cuda:0'), 'validation_loss_bbox': tensor(0.1265, device='cuda:0'), 'validation_loss_giou': tensor(0.4486, device='cuda:0'), 'validation_cardinality_error': tensor(1.2020, device='cuda:0')}
{'training_loss': tensor(0.7547, device='cuda:0'), 'train_loss_ce': tensor(0.1714, device='cuda:0'), 'train_loss_bbox': tensor(0.0385, device='cuda:0'), 'train_loss_giou': tensor(0.1953, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.8844, device='cuda:0'), 'validation_loss_ce': tensor(0.3863, device='cuda:0'), 'validation_loss_bbox': tensor(0.1247, device='cuda:0'), 'validation_loss_giou': tensor(0.4372, device='cuda:0'), 'validation_cardinality_error': tensor(1.0505, device='cuda:0')}
{'training_loss': tensor(0.6149, device='cuda:0'), 'train_loss_ce': tensor(0.0732, device='cuda:0'), 'train_loss_bbox': tensor(0.0420, device='cuda:0'), 'train_loss_giou': tensor(0.1659, device='cuda:0'), 'train_cardinality_error': tensor(1.6000, device='cuda:0'), 'validation_loss': tensor(1.9739, device='cuda:0'), 'validation_loss_ce': tensor(0.4317, device='cuda:0'), 'validation_loss_bbox': tensor(0.1259, device='cuda:0'), 'validation_loss_giou': tensor(0.4563, device='cuda:0'), 'validation_cardinality_error': tensor(0.9495, device='cuda:0')}
{'training_loss': tensor(0.7402, device='cuda:0'), 'train_loss_ce': tensor(0.1470, device='cuda:0'), 'train_loss_bbox': tensor(0.0381, device='cuda:0'), 'train_loss_giou': tensor(0.2013, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.9754, device='cuda:0'), 'validation_loss_ce': tensor(0.3963, device='cuda:0'), 'validation_loss_bbox': tensor(0.1296, device='cuda:0'), 'validation_loss_giou': tensor(0.4657, device='cuda:0'), 'validation_cardinality_error': tensor(0.8485, device='cuda:0')}
{'training_loss': tensor(0.7696, device='cuda:0'), 'train_loss_ce': tensor(0.0860, device='cuda:0'), 'train_loss_bbox': tensor(0.0650, device='cuda:0'), 'train_loss_giou': tensor(0.1794, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(2.0560, device='cuda:0'), 'validation_loss_ce': tensor(0.4615, device='cuda:0'), 'validation_loss_bbox': tensor(0.1319, device='cuda:0'), 'validation_loss_giou': tensor(0.4674, device='cuda:0'), 'validation_cardinality_error': tensor(0.7879, device='cuda:0')}
{'training_loss': tensor(0.6586, device='cuda:0'), 'train_loss_ce': tensor(0.1435, device='cuda:0'), 'train_loss_bbox': tensor(0.0339, device='cuda:0'), 'train_loss_giou': tensor(0.1727, device='cuda:0'), 'train_cardinality_error': tensor(1.2000, device='cuda:0'), 'validation_loss': tensor(1.9738, device='cuda:0'), 'validation_loss_ce': tensor(0.4302, device='cuda:0'), 'validation_loss_bbox': tensor(0.1304, device='cuda:0'), 'validation_loss_giou': tensor(0.4456, device='cuda:0'), 'validation_cardinality_error': tensor(0.6263, device='cuda:0')}
{'training_loss': tensor(0.6021, device='cuda:0'), 'train_loss_ce': tensor(0.1640, device='cuda:0'), 'train_loss_bbox': tensor(0.0302, device='cuda:0'), 'train_loss_giou': tensor(0.1435, device='cuda:0'), 'train_cardinality_error': tensor(2.4000, device='cuda:0'), 'validation_loss': tensor(1.9352, device='cuda:0'), 'validation_loss_ce': tensor(0.3988, device='cuda:0'), 'validation_loss_bbox': tensor(0.1245, device='cuda:0'), 'validation_loss_giou': tensor(0.4569, device='cuda:0'), 'validation_cardinality_error': tensor(1.0808, device='cuda:0')}
{'training_loss': tensor(0.8966, device='cuda:0'), 'train_loss_ce': tensor(0.1546, device='cuda:0'), 'train_loss_bbox': tensor(0.0577, device='cuda:0'), 'train_loss_giou': tensor(0.2268, device='cuda:0'), 'train_cardinality_error': tensor(1.4000, device='cuda:0'), 'validation_loss': tensor(1.9241, device='cuda:0'), 'validation_loss_ce': tensor(0.4078, device='cuda:0'), 'validation_loss_bbox': tensor(0.1259, device='cuda:0'), 'validation_loss_giou': tensor(0.4435, device='cuda:0'), 'validation_cardinality_error': tensor(1.1818, device='cuda:0')}
{'training_loss': tensor(1.0319, device='cuda:0'), 'train_loss_ce': tensor(0.1619, device='cuda:0'), 'train_loss_bbox': tensor(0.0362, device='cuda:0'), 'train_loss_giou': tensor(0.3445, device='cuda:0'), 'train_cardinality_error': tensor(1.8000, device='cuda:0'), 'validation_loss': tensor(2.0069, device='cuda:0'), 'validation_loss_ce': tensor(0.3800, device='cuda:0'), 'validation_loss_bbox': tensor(0.1369, device='cuda:0'), 'validation_loss_giou': tensor(0.4711, device='cuda:0'), 'validation_cardinality_error': tensor(0.7879, device='cuda:0')}
{'training_loss': tensor(0.9223, device='cuda:0'), 'train_loss_ce': tensor(0.1558, device='cuda:0'), 'train_loss_bbox': tensor(0.0659, device='cuda:0'), 'train_loss_giou': tensor(0.2184, device='cuda:0'), 'train_cardinality_error': tensor(2.2000, device='cuda:0'), 'validation_loss': tensor(1.9303, device='cuda:0'), 'validation_loss_ce': tensor(0.3612, device='cuda:0'), 'validation_loss_bbox': tensor(0.1333, device='cuda:0'), 'validation_loss_giou': tensor(0.4514, device='cuda:0'), 'validation_cardinality_error': tensor(0.7778, device='cuda:0')}
{'training_loss': tensor(0.8721, device='cuda:0'), 'train_loss_ce': tensor(0.2239, device='cuda:0'), 'train_loss_bbox': tensor(0.0450, device='cuda:0'), 'train_loss_giou': tensor(0.2117, device='cuda:0'), 'train_cardinality_error': tensor(3., device='cuda:0'), 'validation_loss': tensor(2.0897, device='cuda:0'), 'validation_loss_ce': tensor(0.3954, device='cuda:0'), 'validation_loss_bbox': tensor(0.1372, device='cuda:0'), 'validation_loss_giou': tensor(0.5040, device='cuda:0'), 'validation_cardinality_error': tensor(1.4545, device='cuda:0')}
{'training_loss': tensor(1.1552, device='cuda:0'), 'train_loss_ce': tensor(0.2509, device='cuda:0'), 'train_loss_bbox': tensor(0.0664, device='cuda:0'), 'train_loss_giou': tensor(0.2862, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0082, device='cuda:0'), 'validation_loss_ce': tensor(0.3658, device='cuda:0'), 'validation_loss_bbox': tensor(0.1394, device='cuda:0'), 'validation_loss_giou': tensor(0.4726, device='cuda:0'), 'validation_cardinality_error': tensor(0.8384, device='cuda:0')}
{'training_loss': tensor(0.8142, device='cuda:0'), 'train_loss_ce': tensor(0.1869, device='cuda:0'), 'train_loss_bbox': tensor(0.0506, device='cuda:0'), 'train_loss_giou': tensor(0.1872, device='cuda:0'), 'train_cardinality_error': tensor(2.8000, device='cuda:0'), 'validation_loss': tensor(2.0076, device='cuda:0'), 'validation_loss_ce': tensor(0.3652, device='cuda:0'), 'validation_loss_bbox': tensor(0.1318, device='cuda:0'), 'validation_loss_giou': tensor(0.4916, device='cuda:0'), 'validation_cardinality_error': tensor(1.1010, device='cuda:0')}
{'training_loss': tensor(1.4431, device='cuda:0'), 'train_loss_ce': tensor(0.3380, device='cuda:0'), 'train_loss_bbox': tensor(0.0939, device='cuda:0'), 'train_loss_giou': tensor(0.3177, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9940, device='cuda:0'), 'validation_loss_ce': tensor(0.4066, device='cuda:0'), 'validation_loss_bbox': tensor(0.1247, device='cuda:0'), 'validation_loss_giou': tensor(0.4819, device='cuda:0'), 'validation_cardinality_error': tensor(0.8586, device='cuda:0')}
{'training_loss': tensor(1.0364, device='cuda:0'), 'train_loss_ce': tensor(0.2702, device='cuda:0'), 'train_loss_bbox': tensor(0.0481, device='cuda:0'), 'train_loss_giou': tensor(0.2628, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9175, device='cuda:0'), 'validation_loss_ce': tensor(0.3807, device='cuda:0'), 'validation_loss_bbox': tensor(0.1269, device='cuda:0'), 'validation_loss_giou': tensor(0.4513, device='cuda:0'), 'validation_cardinality_error': tensor(0.8283, device='cuda:0')}
{'training_loss': tensor(1.0160, device='cuda:0'), 'train_loss_ce': tensor(0.1882, device='cuda:0'), 'train_loss_bbox': tensor(0.0697, device='cuda:0'), 'train_loss_giou': tensor(0.2396, device='cuda:0'), 'train_cardinality_error': tensor(1.4000, device='cuda:0'), 'validation_loss': tensor(2.0068, device='cuda:0'), 'validation_loss_ce': tensor(0.3978, device='cuda:0'), 'validation_loss_bbox': tensor(0.1304, device='cuda:0'), 'validation_loss_giou': tensor(0.4785, device='cuda:0'), 'validation_cardinality_error': tensor(0.6667, device='cuda:0')}
{'training_loss': tensor(0.7601, device='cuda:0'), 'train_loss_ce': tensor(0.2106, device='cuda:0'), 'train_loss_bbox': tensor(0.0381, device='cuda:0'), 'train_loss_giou': tensor(0.1795, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.8925, device='cuda:0'), 'validation_loss_ce': tensor(0.3990, device='cuda:0'), 'validation_loss_bbox': tensor(0.1269, device='cuda:0'), 'validation_loss_giou': tensor(0.4295, device='cuda:0'), 'validation_cardinality_error': tensor(0.6970, device='cuda:0')}
{'training_loss': tensor(0.6129, device='cuda:0'), 'train_loss_ce': tensor(0.1430, device='cuda:0'), 'train_loss_bbox': tensor(0.0323, device='cuda:0'), 'train_loss_giou': tensor(0.1541, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9653, device='cuda:0'), 'validation_loss_ce': tensor(0.4163, device='cuda:0'), 'validation_loss_bbox': tensor(0.1287, device='cuda:0'), 'validation_loss_giou': tensor(0.4527, device='cuda:0'), 'validation_cardinality_error': tensor(0.7273, device='cuda:0')}
{'training_loss': tensor(0.6866, device='cuda:0'), 'train_loss_ce': tensor(0.1647, device='cuda:0'), 'train_loss_bbox': tensor(0.0391, device='cuda:0'), 'train_loss_giou': tensor(0.1632, device='cuda:0'), 'train_cardinality_error': tensor(1.2000, device='cuda:0'), 'validation_loss': tensor(1.9499, device='cuda:0'), 'validation_loss_ce': tensor(0.4386, device='cuda:0'), 'validation_loss_bbox': tensor(0.1267, device='cuda:0'), 'validation_loss_giou': tensor(0.4389, device='cuda:0'), 'validation_cardinality_error': tensor(0.9091, device='cuda:0')}
{'training_loss': tensor(0.6045, device='cuda:0'), 'train_loss_ce': tensor(0.0944, device='cuda:0'), 'train_loss_bbox': tensor(0.0349, device='cuda:0'), 'train_loss_giou': tensor(0.1677, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9993, device='cuda:0'), 'validation_loss_ce': tensor(0.4357, device='cuda:0'), 'validation_loss_bbox': tensor(0.1280, device='cuda:0'), 'validation_loss_giou': tensor(0.4617, device='cuda:0'), 'validation_cardinality_error': tensor(0.8889, device='cuda:0')}
{'training_loss': tensor(1.3557, device='cuda:0'), 'train_loss_ce': tensor(0.1851, device='cuda:0'), 'train_loss_bbox': tensor(0.0766, device='cuda:0'), 'train_loss_giou': tensor(0.3937, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.9931, device='cuda:0'), 'validation_loss_ce': tensor(0.4444, device='cuda:0'), 'validation_loss_bbox': tensor(0.1335, device='cuda:0'), 'validation_loss_giou': tensor(0.4407, device='cuda:0'), 'validation_cardinality_error': tensor(0.8687, device='cuda:0')}
{'training_loss': tensor(0.5330, device='cuda:0'), 'train_loss_ce': tensor(0.1816, device='cuda:0'), 'train_loss_bbox': tensor(0.0202, device='cuda:0'), 'train_loss_giou': tensor(0.1252, device='cuda:0'), 'train_cardinality_error': tensor(3., device='cuda:0'), 'validation_loss': tensor(2.0736, device='cuda:0'), 'validation_loss_ce': tensor(0.4412, device='cuda:0'), 'validation_loss_bbox': tensor(0.1370, device='cuda:0'), 'validation_loss_giou': tensor(0.4736, device='cuda:0'), 'validation_cardinality_error': tensor(1.0101, device='cuda:0')}
{'training_loss': tensor(0.8716, device='cuda:0'), 'train_loss_ce': tensor(0.2841, device='cuda:0'), 'train_loss_bbox': tensor(0.0302, device='cuda:0'), 'train_loss_giou': tensor(0.2181, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(2.0201, device='cuda:0'), 'validation_loss_ce': tensor(0.4935, device='cuda:0'), 'validation_loss_bbox': tensor(0.1270, device='cuda:0'), 'validation_loss_giou': tensor(0.4457, device='cuda:0'), 'validation_cardinality_error': tensor(1.3434, device='cuda:0')}
{'training_loss': tensor(0.7222, device='cuda:0'), 'train_loss_ce': tensor(0.2035, device='cuda:0'), 'train_loss_bbox': tensor(0.0457, device='cuda:0'), 'train_loss_giou': tensor(0.1452, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9688, device='cuda:0'), 'validation_loss_ce': tensor(0.4381, device='cuda:0'), 'validation_loss_bbox': tensor(0.1307, device='cuda:0'), 'validation_loss_giou': tensor(0.4387, device='cuda:0'), 'validation_cardinality_error': tensor(0.7475, device='cuda:0')}
{'training_loss': tensor(0.4216, device='cuda:0'), 'train_loss_ce': tensor(0.0884, device='cuda:0'), 'train_loss_bbox': tensor(0.0304, device='cuda:0'), 'train_loss_giou': tensor(0.0907, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(1.9887, device='cuda:0'), 'validation_loss_ce': tensor(0.4895, device='cuda:0'), 'validation_loss_bbox': tensor(0.1241, device='cuda:0'), 'validation_loss_giou': tensor(0.4392, device='cuda:0'), 'validation_cardinality_error': tensor(0.6263, device='cuda:0')}
{'training_loss': tensor(0.6335, device='cuda:0'), 'train_loss_ce': tensor(0.1363, device='cuda:0'), 'train_loss_bbox': tensor(0.0188, device='cuda:0'), 'train_loss_giou': tensor(0.2015, device='cuda:0'), 'train_cardinality_error': tensor(1.6000, device='cuda:0'), 'validation_loss': tensor(1.9909, device='cuda:0'), 'validation_loss_ce': tensor(0.4815, device='cuda:0'), 'validation_loss_bbox': tensor(0.1249, device='cuda:0'), 'validation_loss_giou': tensor(0.4424, device='cuda:0'), 'validation_cardinality_error': tensor(0.7879, device='cuda:0')}
{'training_loss': tensor(0.5883, device='cuda:0'), 'train_loss_ce': tensor(0.0494, device='cuda:0'), 'train_loss_bbox': tensor(0.0409, device='cuda:0'), 'train_loss_giou': tensor(0.1672, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(1.9791, device='cuda:0'), 'validation_loss_ce': tensor(0.4998, device='cuda:0'), 'validation_loss_bbox': tensor(0.1207, device='cuda:0'), 'validation_loss_giou': tensor(0.4379, device='cuda:0'), 'validation_cardinality_error': tensor(0.8586, device='cuda:0')}
{'training_loss': tensor(0.4353, device='cuda:0'), 'train_loss_ce': tensor(0.0519, device='cuda:0'), 'train_loss_bbox': tensor(0.0218, device='cuda:0'), 'train_loss_giou': tensor(0.1373, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(2.0719, device='cuda:0'), 'validation_loss_ce': tensor(0.5327, device='cuda:0'), 'validation_loss_bbox': tensor(0.1285, device='cuda:0'), 'validation_loss_giou': tensor(0.4484, device='cuda:0'), 'validation_cardinality_error': tensor(0.6667, device='cuda:0')}
{'training_loss': tensor(0.6419, device='cuda:0'), 'train_loss_ce': tensor(0.1072, device='cuda:0'), 'train_loss_bbox': tensor(0.0320, device='cuda:0'), 'train_loss_giou': tensor(0.1874, device='cuda:0'), 'train_cardinality_error': tensor(1.4000, device='cuda:0'), 'validation_loss': tensor(2.1160, device='cuda:0'), 'validation_loss_ce': tensor(0.5016, device='cuda:0'), 'validation_loss_bbox': tensor(0.1378, device='cuda:0'), 'validation_loss_giou': tensor(0.4628, device='cuda:0'), 'validation_cardinality_error': tensor(0.8889, device='cuda:0')}
{'training_loss': tensor(0.7383, device='cuda:0'), 'train_loss_ce': tensor(0.0853, device='cuda:0'), 'train_loss_bbox': tensor(0.0365, device='cuda:0'), 'train_loss_giou': tensor(0.2353, device='cuda:0'), 'train_cardinality_error': tensor(1.4000, device='cuda:0'), 'validation_loss': tensor(2.0324, device='cuda:0'), 'validation_loss_ce': tensor(0.5191, device='cuda:0'), 'validation_loss_bbox': tensor(0.1237, device='cuda:0'), 'validation_loss_giou': tensor(0.4473, device='cuda:0'), 'validation_cardinality_error': tensor(1.0909, device='cuda:0')}
{'training_loss': tensor(0.7048, device='cuda:0'), 'train_loss_ce': tensor(0.1198, device='cuda:0'), 'train_loss_bbox': tensor(0.0352, device='cuda:0'), 'train_loss_giou': tensor(0.2044, device='cuda:0'), 'train_cardinality_error': tensor(1.4000, device='cuda:0'), 'validation_loss': tensor(2.0031, device='cuda:0'), 'validation_loss_ce': tensor(0.5021, device='cuda:0'), 'validation_loss_bbox': tensor(0.1242, device='cuda:0'), 'validation_loss_giou': tensor(0.4400, device='cuda:0'), 'validation_cardinality_error': tensor(0.8586, device='cuda:0')}
{'training_loss': tensor(0.7183, device='cuda:0'), 'train_loss_ce': tensor(0.0303, device='cuda:0'), 'train_loss_bbox': tensor(0.0404, device='cuda:0'), 'train_loss_giou': tensor(0.2429, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(2.0240, device='cuda:0'), 'validation_loss_ce': tensor(0.5325, device='cuda:0'), 'validation_loss_bbox': tensor(0.1234, device='cuda:0'), 'validation_loss_giou': tensor(0.4374, device='cuda:0'), 'validation_cardinality_error': tensor(0.8990, device='cuda:0')}
{'training_loss': tensor(0.5812, device='cuda:0'), 'train_loss_ce': tensor(0.0670, device='cuda:0'), 'train_loss_bbox': tensor(0.0321, device='cuda:0'), 'train_loss_giou': tensor(0.1768, device='cuda:0'), 'train_cardinality_error': tensor(1.2000, device='cuda:0'), 'validation_loss': tensor(2.0031, device='cuda:0'), 'validation_loss_ce': tensor(0.5461, device='cuda:0'), 'validation_loss_bbox': tensor(0.1206, device='cuda:0'), 'validation_loss_giou': tensor(0.4271, device='cuda:0'), 'validation_cardinality_error': tensor(0.7172, device='cuda:0')}
{'training_loss': tensor(0.4142, device='cuda:0'), 'train_loss_ce': tensor(0.0514, device='cuda:0'), 'train_loss_bbox': tensor(0.0342, device='cuda:0'), 'train_loss_giou': tensor(0.0958, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0898, device='cuda:0'), 'validation_loss_ce': tensor(0.5431, device='cuda:0'), 'validation_loss_bbox': tensor(0.1331, device='cuda:0'), 'validation_loss_giou': tensor(0.4406, device='cuda:0'), 'validation_cardinality_error': tensor(0.8283, device='cuda:0')}
{'training_loss': tensor(0.5866, device='cuda:0'), 'train_loss_ce': tensor(0.0849, device='cuda:0'), 'train_loss_bbox': tensor(0.0300, device='cuda:0'), 'train_loss_giou': tensor(0.1758, device='cuda:0'), 'train_cardinality_error': tensor(2.6000, device='cuda:0'), 'validation_loss': tensor(2.0080, device='cuda:0'), 'validation_loss_ce': tensor(0.4937, device='cuda:0'), 'validation_loss_bbox': tensor(0.1242, device='cuda:0'), 'validation_loss_giou': tensor(0.4467, device='cuda:0'), 'validation_cardinality_error': tensor(0.6465, device='cuda:0')}
{'training_loss': tensor(0.5337, device='cuda:0'), 'train_loss_ce': tensor(0.0545, device='cuda:0'), 'train_loss_bbox': tensor(0.0357, device='cuda:0'), 'train_loss_giou': tensor(0.1503, device='cuda:0'), 'train_cardinality_error': tensor(0.2000, device='cuda:0'), 'validation_loss': tensor(2.1421, device='cuda:0'), 'validation_loss_ce': tensor(0.5116, device='cuda:0'), 'validation_loss_bbox': tensor(0.1371, device='cuda:0'), 'validation_loss_giou': tensor(0.4725, device='cuda:0'), 'validation_cardinality_error': tensor(0.9192, device='cuda:0')}
{'training_loss': tensor(0.7298, device='cuda:0'), 'train_loss_ce': tensor(0.0942, device='cuda:0'), 'train_loss_bbox': tensor(0.0284, device='cuda:0'), 'train_loss_giou': tensor(0.2468, device='cuda:0'), 'train_cardinality_error': tensor(2.4000, device='cuda:0'), 'validation_loss': tensor(2.1484, device='cuda:0'), 'validation_loss_ce': tensor(0.5031, device='cuda:0'), 'validation_loss_bbox': tensor(0.1347, device='cuda:0'), 'validation_loss_giou': tensor(0.4860, device='cuda:0'), 'validation_cardinality_error': tensor(0.6768, device='cuda:0')}
{'training_loss': tensor(0.8502, device='cuda:0'), 'train_loss_ce': tensor(0.0793, device='cuda:0'), 'train_loss_bbox': tensor(0.0461, device='cuda:0'), 'train_loss_giou': tensor(0.2702, device='cuda:0'), 'train_cardinality_error': tensor(0.6000, device='cuda:0'), 'validation_loss': tensor(2.1865, device='cuda:0'), 'validation_loss_ce': tensor(0.5221, device='cuda:0'), 'validation_loss_bbox': tensor(0.1374, device='cuda:0'), 'validation_loss_giou': tensor(0.4888, device='cuda:0'), 'validation_cardinality_error': tensor(0.6667, device='cuda:0')}
{'training_loss': tensor(0.5967, device='cuda:0'), 'train_loss_ce': tensor(0.1064, device='cuda:0'), 'train_loss_bbox': tensor(0.0283, device='cuda:0'), 'train_loss_giou': tensor(0.1744, device='cuda:0'), 'train_cardinality_error': tensor(0.4000, device='cuda:0'), 'validation_loss': tensor(2.1303, device='cuda:0'), 'validation_loss_ce': tensor(0.5539, device='cuda:0'), 'validation_loss_bbox': tensor(0.1321, device='cuda:0'), 'validation_loss_giou': tensor(0.4579, device='cuda:0'), 'validation_cardinality_error': tensor(0.6263, device='cuda:0')}
{'training_loss': tensor(0.6167, device='cuda:0'), 'train_loss_ce': tensor(0.1510, device='cuda:0'), 'train_loss_bbox': tensor(0.0336, device='cuda:0'), 'train_loss_giou': tensor(0.1490, device='cuda:0'), 'train_cardinality_error': tensor(2., device='cuda:0'), 'validation_loss': tensor(2.1606, device='cuda:0'), 'validation_loss_ce': tensor(0.5449, device='cuda:0'), 'validation_loss_bbox': tensor(0.1357, device='cuda:0'), 'validation_loss_giou': tensor(0.4686, device='cuda:0'), 'validation_cardinality_error': tensor(0.8081, device='cuda:0')}
```

## Examples
{'size': tensor([512, 512]), 'image_id': tensor([1]), 'class_labels': tensor([4]), 'boxes': tensor([[0.2622, 0.5729, 0.0847, 0.0773]]), 'area': tensor([1717.9431]), 'iscrowd': tensor([0]), 'orig_size': tensor([1024, 1024])}

![Example](./example.png)