Update README file

README.md

@@ -13,34 +13,34 @@ IoU metric: bbox
  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 ] = 0.000
- Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 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 ] = 0.008
  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 ] = 0.000
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.008
 ```
 
 ## After training 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 ] = 0.001
- Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.000
+ Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.008
+ Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.016
+ Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.009
  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 ] = 0.001
- Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.019
- Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.021
+ Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.008
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.060
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.122
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.140
  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 ] = 0.021
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.141
 ```
 
 ## Config
 - dataset: NIH
 - original model: hustvl/yolos-tiny
-- lr: 1e-06
+- lr: 1e-05
 - dropout_rate: 0.1
 - weight_decay: 0.05
 - max_epochs: 50
@@ -49,57 +49,57 @@ IoU metric: bbox
 ## Logging
 ### Training process
 ```
-{'validation_loss': tensor(7.1565, device='cuda:0'), 'validation_loss_ce': tensor(2.6168, device='cuda:0'), 'validation_loss_bbox': tensor(0.5211, device='cuda:0'), 'validation_loss_giou': tensor(0.9671, device='cuda:0'), 'validation_cardinality_error': tensor(99., device='cuda:0')}
-{'training_loss': tensor(6.0705, device='cuda:0'), 'train_loss_ce': tensor(2.5302, device='cuda:0'), 'train_loss_bbox': tensor(0.4304, device='cuda:0'), 'train_loss_giou': tensor(0.6943, device='cuda:0'), 'train_cardinality_error': tensor(99., device='cuda:0'), 'validation_loss': tensor(6.3058, device='cuda:0'), 'validation_loss_ce': tensor(2.4960, device='cuda:0'), 'validation_loss_bbox': tensor(0.4287, device='cuda:0'), 'validation_loss_giou': tensor(0.8332, device='cuda:0'), 'validation_cardinality_error': tensor(98.8889, device='cuda:0')}
-{'training_loss': tensor(5.3459, device='cuda:0'), 'train_loss_ce': tensor(2.4056, device='cuda:0'), 'train_loss_bbox': tensor(0.3181, device='cuda:0'), 'train_loss_giou': tensor(0.6750, device='cuda:0'), 'train_cardinality_error': tensor(98.6000, device='cuda:0'), 'validation_loss': tensor(5.6501, device='cuda:0'), 'validation_loss_ce': tensor(2.3928, device='cuda:0'), 'validation_loss_bbox': tensor(0.3429, device='cuda:0'), 'validation_loss_giou': tensor(0.7713, device='cuda:0'), 'validation_cardinality_error': tensor(98.2525, device='cuda:0')}
-{'training_loss': tensor(5.3122, device='cuda:0'), 'train_loss_ce': tensor(2.2708, device='cuda:0'), 'train_loss_bbox': tensor(0.3198, device='cuda:0'), 'train_loss_giou': tensor(0.7211, device='cuda:0'), 'train_cardinality_error': tensor(97.8000, device='cuda:0'), 'validation_loss': tensor(5.2498, device='cuda:0'), 'validation_loss_ce': tensor(2.2861, device='cuda:0'), 'validation_loss_bbox': tensor(0.3017, device='cuda:0'), 'validation_loss_giou': tensor(0.7276, device='cuda:0'), 'validation_cardinality_error': tensor(96.0101, device='cuda:0')}
-{'training_loss': tensor(3.9929, device='cuda:0'), 'train_loss_ce': tensor(2.1223, device='cuda:0'), 'train_loss_bbox': tensor(0.2062, device='cuda:0'), 'train_loss_giou': tensor(0.4198, device='cuda:0'), 'train_cardinality_error': tensor(84., device='cuda:0'), 'validation_loss': tensor(5.0342, device='cuda:0'), 'validation_loss_ce': tensor(2.1912, device='cuda:0'), 'validation_loss_bbox': tensor(0.2814, device='cuda:0'), 'validation_loss_giou': tensor(0.7179, device='cuda:0'), 'validation_cardinality_error': tensor(91., device='cuda:0')}
-{'training_loss': tensor(4.1406, device='cuda:0'), 'train_loss_ce': tensor(2.1059, device='cuda:0'), 'train_loss_bbox': tensor(0.2164, device='cuda:0'), 'train_loss_giou': tensor(0.4762, device='cuda:0'), 'train_cardinality_error': tensor(84., device='cuda:0'), 'validation_loss': tensor(4.8572, device='cuda:0'), 'validation_loss_ce': tensor(2.0894, device='cuda:0'), 'validation_loss_bbox': tensor(0.2727, device='cuda:0'), 'validation_loss_giou': tensor(0.7021, device='cuda:0'), 'validation_cardinality_error': tensor(78.2929, device='cuda:0')}
-{'training_loss': tensor(4.5332, device='cuda:0'), 'train_loss_ce': tensor(1.9598, device='cuda:0'), 'train_loss_bbox': tensor(0.2270, device='cuda:0'), 'train_loss_giou': tensor(0.7191, device='cuda:0'), 'train_cardinality_error': tensor(50.2000, device='cuda:0'), 'validation_loss': tensor(4.6922, device='cuda:0'), 'validation_loss_ce': tensor(1.9838, device='cuda:0'), 'validation_loss_bbox': tensor(0.2630, device='cuda:0'), 'validation_loss_giou': tensor(0.6967, device='cuda:0'), 'validation_cardinality_error': tensor(55.5758, device='cuda:0')}
-{'training_loss': tensor(4.4722, device='cuda:0'), 'train_loss_ce': tensor(1.7903, device='cuda:0'), 'train_loss_bbox': tensor(0.2521, device='cuda:0'), 'train_loss_giou': tensor(0.7108, device='cuda:0'), 'train_cardinality_error': tensor(8.2000, device='cuda:0'), 'validation_loss': tensor(4.5388, device='cuda:0'), 'validation_loss_ce': tensor(1.8704, device='cuda:0'), 'validation_loss_bbox': tensor(0.2569, device='cuda:0'), 'validation_loss_giou': tensor(0.6921, device='cuda:0'), 'validation_cardinality_error': tensor(30.3838, device='cuda:0')}
-{'training_loss': tensor(4.5265, device='cuda:0'), 'train_loss_ce': tensor(1.6432, device='cuda:0'), 'train_loss_bbox': tensor(0.2652, device='cuda:0'), 'train_loss_giou': tensor(0.7785, device='cuda:0'), 'train_cardinality_error': tensor(3., device='cuda:0'), 'validation_loss': tensor(4.3567, device='cuda:0'), 'validation_loss_ce': tensor(1.7606, device='cuda:0'), 'validation_loss_bbox': tensor(0.2432, device='cuda:0'), 'validation_loss_giou': tensor(0.6902, device='cuda:0'), 'validation_cardinality_error': tensor(13.5960, device='cuda:0')}
-{'training_loss': tensor(4.7480, device='cuda:0'), 'train_loss_ce': tensor(1.6500, device='cuda:0'), 'train_loss_bbox': tensor(0.2890, device='cuda:0'), 'train_loss_giou': tensor(0.8264, device='cuda:0'), 'train_cardinality_error': tensor(5., device='cuda:0'), 'validation_loss': tensor(4.2126, device='cuda:0'), 'validation_loss_ce': tensor(1.6448, device='cuda:0'), 'validation_loss_bbox': tensor(0.2399, device='cuda:0'), 'validation_loss_giou': tensor(0.6841, device='cuda:0'), 'validation_cardinality_error': tensor(4.7879, device='cuda:0')}
-{'training_loss': tensor(3.8881, device='cuda:0'), 'train_loss_ce': tensor(1.5876, device='cuda:0'), 'train_loss_bbox': tensor(0.2124, device='cuda:0'), 'train_loss_giou': tensor(0.6194, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(4.0765, device='cuda:0'), 'validation_loss_ce': tensor(1.5303, device='cuda:0'), 'validation_loss_bbox': tensor(0.2369, device='cuda:0'), 'validation_loss_giou': tensor(0.6807, device='cuda:0'), 'validation_cardinality_error': tensor(1.6667, device='cuda:0')}
-{'training_loss': tensor(4.0165, device='cuda:0'), 'train_loss_ce': tensor(1.3808, device='cuda:0'), 'train_loss_bbox': tensor(0.2712, device='cuda:0'), 'train_loss_giou': tensor(0.6398, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.9377, device='cuda:0'), 'validation_loss_ce': tensor(1.4131, device='cuda:0'), 'validation_loss_bbox': tensor(0.2350, device='cuda:0'), 'validation_loss_giou': tensor(0.6749, device='cuda:0'), 'validation_cardinality_error': tensor(0.9697, device='cuda:0')}
-{'training_loss': tensor(4.0245, device='cuda:0'), 'train_loss_ce': tensor(1.2671, device='cuda:0'), 'train_loss_bbox': tensor(0.2208, device='cuda:0'), 'train_loss_giou': tensor(0.8266, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.7913, device='cuda:0'), 'validation_loss_ce': tensor(1.2989, device='cuda:0'), 'validation_loss_bbox': tensor(0.2306, device='cuda:0'), 'validation_loss_giou': tensor(0.6698, device='cuda:0'), 'validation_cardinality_error': tensor(0.9697, device='cuda:0')}
-{'training_loss': tensor(3.4285, device='cuda:0'), 'train_loss_ce': tensor(1.1165, device='cuda:0'), 'train_loss_bbox': tensor(0.1885, device='cuda:0'), 'train_loss_giou': tensor(0.6848, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.6641, device='cuda:0'), 'validation_loss_ce': tensor(1.1870, device='cuda:0'), 'validation_loss_bbox': tensor(0.2271, device='cuda:0'), 'validation_loss_giou': tensor(0.6708, device='cuda:0'), 'validation_cardinality_error': tensor(0.9798, device='cuda:0')}
-{'training_loss': tensor(3.7639, device='cuda:0'), 'train_loss_ce': tensor(1.1571, device='cuda:0'), 'train_loss_bbox': tensor(0.2647, device='cuda:0'), 'train_loss_giou': tensor(0.6416, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.5421, device='cuda:0'), 'validation_loss_ce': tensor(1.0872, device='cuda:0'), 'validation_loss_bbox': tensor(0.2242, device='cuda:0'), 'validation_loss_giou': tensor(0.6668, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.9899, device='cuda:0'), 'train_loss_ce': tensor(0.9906, device='cuda:0'), 'train_loss_bbox': tensor(0.1988, device='cuda:0'), 'train_loss_giou': tensor(0.5028, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.4229, device='cuda:0'), 'validation_loss_ce': tensor(0.9928, device='cuda:0'), 'validation_loss_bbox': tensor(0.2208, device='cuda:0'), 'validation_loss_giou': tensor(0.6629, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.8417, device='cuda:0'), 'train_loss_ce': tensor(0.9287, device='cuda:0'), 'train_loss_bbox': tensor(0.1494, device='cuda:0'), 'train_loss_giou': tensor(0.5831, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.3267, device='cuda:0'), 'validation_loss_ce': tensor(0.9071, device='cuda:0'), 'validation_loss_bbox': tensor(0.2219, device='cuda:0'), 'validation_loss_giou': tensor(0.6550, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.9143, device='cuda:0'), 'train_loss_ce': tensor(0.8071, device='cuda:0'), 'train_loss_bbox': tensor(0.1854, device='cuda:0'), 'train_loss_giou': tensor(0.5901, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.2317, device='cuda:0'), 'validation_loss_ce': tensor(0.8366, device='cuda:0'), 'validation_loss_bbox': tensor(0.2197, device='cuda:0'), 'validation_loss_giou': tensor(0.6482, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.2213, device='cuda:0'), 'train_loss_ce': tensor(0.8347, device='cuda:0'), 'train_loss_bbox': tensor(0.2265, device='cuda:0'), 'train_loss_giou': tensor(0.6270, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.1624, device='cuda:0'), 'validation_loss_ce': tensor(0.7770, device='cuda:0'), 'validation_loss_bbox': tensor(0.2183, device='cuda:0'), 'validation_loss_giou': tensor(0.6469, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.1193, device='cuda:0'), 'train_loss_ce': tensor(0.6932, device='cuda:0'), 'train_loss_bbox': tensor(0.1298, device='cuda:0'), 'train_loss_giou': tensor(0.3885, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.0898, device='cuda:0'), 'validation_loss_ce': tensor(0.7267, device='cuda:0'), 'validation_loss_bbox': tensor(0.2196, device='cuda:0'), 'validation_loss_giou': tensor(0.6326, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.1823, device='cuda:0'), 'train_loss_ce': tensor(0.6994, device='cuda:0'), 'train_loss_bbox': tensor(0.2211, device='cuda:0'), 'train_loss_giou': tensor(0.6887, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.0247, device='cuda:0'), 'validation_loss_ce': tensor(0.6842, device='cuda:0'), 'validation_loss_bbox': tensor(0.2153, device='cuda:0'), 'validation_loss_giou': tensor(0.6321, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.7933, device='cuda:0'), 'train_loss_ce': tensor(0.6156, device='cuda:0'), 'train_loss_bbox': tensor(0.1934, device='cuda:0'), 'train_loss_giou': tensor(0.6054, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.9765, device='cuda:0'), 'validation_loss_ce': tensor(0.6504, device='cuda:0'), 'validation_loss_bbox': tensor(0.2157, device='cuda:0'), 'validation_loss_giou': tensor(0.6237, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.2350, device='cuda:0'), 'train_loss_ce': tensor(0.6573, device='cuda:0'), 'train_loss_bbox': tensor(0.2416, device='cuda:0'), 'train_loss_giou': tensor(0.6849, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.9092, device='cuda:0'), 'validation_loss_ce': tensor(0.6215, device='cuda:0'), 'validation_loss_bbox': tensor(0.2121, device='cuda:0'), 'validation_loss_giou': tensor(0.6135, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.0423, device='cuda:0'), 'train_loss_ce': tensor(0.6132, device='cuda:0'), 'train_loss_bbox': tensor(0.1892, device='cuda:0'), 'train_loss_giou': tensor(0.7415, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.9006, device='cuda:0'), 'validation_loss_ce': tensor(0.5973, device='cuda:0'), 'validation_loss_bbox': tensor(0.2138, device='cuda:0'), 'validation_loss_giou': tensor(0.6170, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.2000, device='cuda:0'), 'train_loss_ce': tensor(0.5179, device='cuda:0'), 'train_loss_bbox': tensor(0.1494, device='cuda:0'), 'train_loss_giou': tensor(0.4675, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.8636, device='cuda:0'), 'validation_loss_ce': tensor(0.5780, device='cuda:0'), 'validation_loss_bbox': tensor(0.2105, device='cuda:0'), 'validation_loss_giou': tensor(0.6167, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.4896, device='cuda:0'), 'train_loss_ce': tensor(0.5899, device='cuda:0'), 'train_loss_bbox': tensor(0.2030, device='cuda:0'), 'train_loss_giou': tensor(0.4422, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.8427, device='cuda:0'), 'validation_loss_ce': tensor(0.5624, device='cuda:0'), 'validation_loss_bbox': tensor(0.2102, device='cuda:0'), 'validation_loss_giou': tensor(0.6146, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.3007, device='cuda:0'), 'train_loss_ce': tensor(0.4799, device='cuda:0'), 'train_loss_bbox': tensor(0.1701, device='cuda:0'), 'train_loss_giou': tensor(0.4853, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.8161, device='cuda:0'), 'validation_loss_ce': tensor(0.5492, device='cuda:0'), 'validation_loss_bbox': tensor(0.2090, device='cuda:0'), 'validation_loss_giou': tensor(0.6110, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.4182, device='cuda:0'), 'train_loss_ce': tensor(0.5711, device='cuda:0'), 'train_loss_bbox': tensor(0.1410, device='cuda:0'), 'train_loss_giou': tensor(0.5709, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.8062, device='cuda:0'), 'validation_loss_ce': tensor(0.5394, device='cuda:0'), 'validation_loss_bbox': tensor(0.2077, device='cuda:0'), 'validation_loss_giou': tensor(0.6141, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.0376, device='cuda:0'), 'train_loss_ce': tensor(0.5342, device='cuda:0'), 'train_loss_bbox': tensor(0.1438, device='cuda:0'), 'train_loss_giou': tensor(0.3921, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7929, device='cuda:0'), 'validation_loss_ce': tensor(0.5284, device='cuda:0'), 'validation_loss_bbox': tensor(0.2054, device='cuda:0'), 'validation_loss_giou': tensor(0.6187, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(1.9113, device='cuda:0'), 'train_loss_ce': tensor(0.5603, device='cuda:0'), 'train_loss_bbox': tensor(0.1336, device='cuda:0'), 'train_loss_giou': tensor(0.3416, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7774, device='cuda:0'), 'validation_loss_ce': tensor(0.5212, device='cuda:0'), 'validation_loss_bbox': tensor(0.2079, device='cuda:0'), 'validation_loss_giou': tensor(0.6084, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(1.9175, device='cuda:0'), 'train_loss_ce': tensor(0.5524, device='cuda:0'), 'train_loss_bbox': tensor(0.1297, device='cuda:0'), 'train_loss_giou': tensor(0.3582, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7511, device='cuda:0'), 'validation_loss_ce': tensor(0.5131, device='cuda:0'), 'validation_loss_bbox': tensor(0.2043, device='cuda:0'), 'validation_loss_giou': tensor(0.6082, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.0097, device='cuda:0'), 'train_loss_ce': tensor(0.4589, device='cuda:0'), 'train_loss_bbox': tensor(0.1780, device='cuda:0'), 'train_loss_giou': tensor(0.8303, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7464, device='cuda:0'), 'validation_loss_ce': tensor(0.5085, device='cuda:0'), 'validation_loss_bbox': tensor(0.2042, device='cuda:0'), 'validation_loss_giou': tensor(0.6085, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(1.6472, device='cuda:0'), 'train_loss_ce': tensor(0.5455, device='cuda:0'), 'train_loss_bbox': tensor(0.1061, device='cuda:0'), 'train_loss_giou': tensor(0.2856, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7289, device='cuda:0'), 'validation_loss_ce': tensor(0.5024, device='cuda:0'), 'validation_loss_bbox': tensor(0.2017, device='cuda:0'), 'validation_loss_giou': tensor(0.6090, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.2923, device='cuda:0'), 'train_loss_ce': tensor(0.5256, device='cuda:0'), 'train_loss_bbox': tensor(0.1532, device='cuda:0'), 'train_loss_giou': tensor(0.5003, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7224, device='cuda:0'), 'validation_loss_ce': tensor(0.4985, device='cuda:0'), 'validation_loss_bbox': tensor(0.2026, device='cuda:0'), 'validation_loss_giou': tensor(0.6054, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.0831, device='cuda:0'), 'train_loss_ce': tensor(0.5228, device='cuda:0'), 'train_loss_bbox': tensor(0.1292, device='cuda:0'), 'train_loss_giou': tensor(0.4571, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7133, device='cuda:0'), 'validation_loss_ce': tensor(0.4948, device='cuda:0'), 'validation_loss_bbox': tensor(0.2020, device='cuda:0'), 'validation_loss_giou': tensor(0.6043, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.8135, device='cuda:0'), 'train_loss_ce': tensor(0.4272, device='cuda:0'), 'train_loss_bbox': tensor(0.2240, device='cuda:0'), 'train_loss_giou': tensor(0.6332, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7171, device='cuda:0'), 'validation_loss_ce': tensor(0.4908, device='cuda:0'), 'validation_loss_bbox': tensor(0.2027, device='cuda:0'), 'validation_loss_giou': tensor(0.6064, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.2130, device='cuda:0'), 'train_loss_ce': tensor(0.4784, device='cuda:0'), 'train_loss_bbox': tensor(0.1390, device='cuda:0'), 'train_loss_giou': tensor(0.5198, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6983, device='cuda:0'), 'validation_loss_ce': tensor(0.4877, device='cuda:0'), 'validation_loss_bbox': tensor(0.2003, device='cuda:0'), 'validation_loss_giou': tensor(0.6044, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.7309, device='cuda:0'), 'train_loss_ce': tensor(0.5104, device='cuda:0'), 'train_loss_bbox': tensor(0.1930, device='cuda:0'), 'train_loss_giou': tensor(0.6278, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6918, device='cuda:0'), 'validation_loss_ce': tensor(0.4846, device='cuda:0'), 'validation_loss_bbox': tensor(0.2006, device='cuda:0'), 'validation_loss_giou': tensor(0.6022, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(1.6741, device='cuda:0'), 'train_loss_ce': tensor(0.5058, device='cuda:0'), 'train_loss_bbox': tensor(0.0880, device='cuda:0'), 'train_loss_giou': tensor(0.3642, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6956, device='cuda:0'), 'validation_loss_ce': tensor(0.4832, device='cuda:0'), 'validation_loss_bbox': tensor(0.2020, device='cuda:0'), 'validation_loss_giou': tensor(0.6012, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.0059, device='cuda:0'), 'train_loss_ce': tensor(0.4803, device='cuda:0'), 'train_loss_bbox': tensor(0.2021, device='cuda:0'), 'train_loss_giou': tensor(0.7575, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6876, device='cuda:0'), 'validation_loss_ce': tensor(0.4818, device='cuda:0'), 'validation_loss_bbox': tensor(0.2015, device='cuda:0'), 'validation_loss_giou': tensor(0.5990, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.7421, device='cuda:0'), 'train_loss_ce': tensor(0.5130, device='cuda:0'), 'train_loss_bbox': tensor(0.1896, device='cuda:0'), 'train_loss_giou': tensor(0.6407, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6769, device='cuda:0'), 'validation_loss_ce': tensor(0.4791, device='cuda:0'), 'validation_loss_bbox': tensor(0.2004, device='cuda:0'), 'validation_loss_giou': tensor(0.5979, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(1.7125, device='cuda:0'), 'train_loss_ce': tensor(0.4362, device='cuda:0'), 'train_loss_bbox': tensor(0.1277, device='cuda:0'), 'train_loss_giou': tensor(0.3188, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6743, device='cuda:0'), 'validation_loss_ce': tensor(0.4768, device='cuda:0'), 'validation_loss_bbox': tensor(0.1980, device='cuda:0'), 'validation_loss_giou': tensor(0.6038, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.2567, device='cuda:0'), 'train_loss_ce': tensor(0.5142, device='cuda:0'), 'train_loss_bbox': tensor(0.1576, device='cuda:0'), 'train_loss_giou': tensor(0.4773, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6748, device='cuda:0'), 'validation_loss_ce': tensor(0.4765, device='cuda:0'), 'validation_loss_bbox': tensor(0.2005, device='cuda:0'), 'validation_loss_giou': tensor(0.5980, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(1.5372, device='cuda:0'), 'train_loss_ce': tensor(0.5373, device='cuda:0'), 'train_loss_bbox': tensor(0.0994, device='cuda:0'), 'train_loss_giou': tensor(0.2516, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6640, device='cuda:0'), 'validation_loss_ce': tensor(0.4742, device='cuda:0'), 'validation_loss_bbox': tensor(0.1982, device='cuda:0'), 'validation_loss_giou': tensor(0.5993, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.5040, device='cuda:0'), 'train_loss_ce': tensor(0.4272, device='cuda:0'), 'train_loss_bbox': tensor(0.1789, device='cuda:0'), 'train_loss_giou': tensor(0.5912, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6623, device='cuda:0'), 'validation_loss_ce': tensor(0.4728, device='cuda:0'), 'validation_loss_bbox': tensor(0.1981, device='cuda:0'), 'validation_loss_giou': tensor(0.5996, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.0120, device='cuda:0'), 'train_loss_ce': tensor(0.4059, device='cuda:0'), 'train_loss_bbox': tensor(0.1301, device='cuda:0'), 'train_loss_giou': tensor(0.4778, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6473, device='cuda:0'), 'validation_loss_ce': tensor(0.4709, device='cuda:0'), 'validation_loss_bbox': tensor(0.1967, device='cuda:0'), 'validation_loss_giou': tensor(0.5964, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.0679, device='cuda:0'), 'train_loss_ce': tensor(0.5134, device='cuda:0'), 'train_loss_bbox': tensor(0.1250, device='cuda:0'), 'train_loss_giou': tensor(0.4649, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6562, device='cuda:0'), 'validation_loss_ce': tensor(0.4703, device='cuda:0'), 'validation_loss_bbox': tensor(0.1983, device='cuda:0'), 'validation_loss_giou': tensor(0.5971, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.2338, device='cuda:0'), 'train_loss_ce': tensor(0.5023, device='cuda:0'), 'train_loss_bbox': tensor(0.1235, device='cuda:0'), 'train_loss_giou': tensor(0.5569, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6502, device='cuda:0'), 'validation_loss_ce': tensor(0.4681, device='cuda:0'), 'validation_loss_bbox': tensor(0.1989, device='cuda:0'), 'validation_loss_giou': tensor(0.5938, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(3.3818, device='cuda:0'), 'train_loss_ce': tensor(0.2822, device='cuda:0'), 'train_loss_bbox': tensor(0.2465, device='cuda:0'), 'train_loss_giou': tensor(0.9336, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6623, device='cuda:0'), 'validation_loss_ce': tensor(0.4676, device='cuda:0'), 'validation_loss_bbox': tensor(0.1998, device='cuda:0'), 'validation_loss_giou': tensor(0.5978, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.7525, device='cuda:0'), 'train_loss_ce': tensor(0.4899, device='cuda:0'), 'train_loss_bbox': tensor(0.1936, device='cuda:0'), 'train_loss_giou': tensor(0.6474, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6486, device='cuda:0'), 'validation_loss_ce': tensor(0.4659, device='cuda:0'), 'validation_loss_bbox': tensor(0.1986, device='cuda:0'), 'validation_loss_giou': tensor(0.5948, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
-{'training_loss': tensor(2.1041, device='cuda:0'), 'train_loss_ce': tensor(0.4949, device='cuda:0'), 'train_loss_bbox': tensor(0.1062, device='cuda:0'), 'train_loss_giou': tensor(0.5390, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6575, device='cuda:0'), 'validation_loss_ce': tensor(0.4652, device='cuda:0'), 'validation_loss_bbox': tensor(0.2000, device='cuda:0'), 'validation_loss_giou': tensor(0.5961, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'validation_loss': tensor(7.0099, device='cuda:0'), 'validation_loss_ce': tensor(2.4436, device='cuda:0'), 'validation_loss_bbox': tensor(0.5540, device='cuda:0'), 'validation_loss_giou': tensor(0.8981, device='cuda:0'), 'validation_cardinality_error': tensor(99., device='cuda:0')}
+{'training_loss': tensor(3.9420, device='cuda:0'), 'train_loss_ce': tensor(1.3197, device='cuda:0'), 'train_loss_bbox': tensor(0.2500, device='cuda:0'), 'train_loss_giou': tensor(0.6860, device='cuda:0'), 'train_cardinality_error': tensor(7., device='cuda:0'), 'validation_loss': tensor(3.3165, device='cuda:0'), 'validation_loss_ce': tensor(1.2489, device='cuda:0'), 'validation_loss_bbox': tensor(0.1881, device='cuda:0'), 'validation_loss_giou': tensor(0.5635, device='cuda:0'), 'validation_cardinality_error': tensor(4.0101, device='cuda:0')}
+{'training_loss': tensor(3.3235, device='cuda:0'), 'train_loss_ce': tensor(0.6793, device='cuda:0'), 'train_loss_bbox': tensor(0.2541, device='cuda:0'), 'train_loss_giou': tensor(0.6867, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.4833, device='cuda:0'), 'validation_loss_ce': tensor(0.6021, device='cuda:0'), 'validation_loss_bbox': tensor(0.1646, device='cuda:0'), 'validation_loss_giou': tensor(0.5291, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.3983, device='cuda:0'), 'train_loss_ce': tensor(0.5598, device='cuda:0'), 'train_loss_bbox': tensor(0.1766, device='cuda:0'), 'train_loss_giou': tensor(0.4778, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2504, device='cuda:0'), 'validation_loss_ce': tensor(0.4912, device='cuda:0'), 'validation_loss_bbox': tensor(0.1561, device='cuda:0'), 'validation_loss_giou': tensor(0.4895, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.9965, device='cuda:0'), 'train_loss_ce': tensor(0.4444, device='cuda:0'), 'train_loss_bbox': tensor(0.1335, device='cuda:0'), 'train_loss_giou': tensor(0.4423, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2376, device='cuda:0'), 'validation_loss_ce': tensor(0.4733, device='cuda:0'), 'validation_loss_bbox': tensor(0.1545, device='cuda:0'), 'validation_loss_giou': tensor(0.4960, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.3605, device='cuda:0'), 'train_loss_ce': tensor(0.3837, device='cuda:0'), 'train_loss_bbox': tensor(0.1640, device='cuda:0'), 'train_loss_giou': tensor(0.5785, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2257, device='cuda:0'), 'validation_loss_ce': tensor(0.4630, device='cuda:0'), 'validation_loss_bbox': tensor(0.1510, device='cuda:0'), 'validation_loss_giou': tensor(0.5039, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.8496, device='cuda:0'), 'train_loss_ce': tensor(0.4450, device='cuda:0'), 'train_loss_bbox': tensor(0.1345, device='cuda:0'), 'train_loss_giou': tensor(0.3660, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0929, device='cuda:0'), 'validation_loss_ce': tensor(0.4594, device='cuda:0'), 'validation_loss_bbox': tensor(0.1348, device='cuda:0'), 'validation_loss_giou': tensor(0.4796, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.8037, device='cuda:0'), 'train_loss_ce': tensor(0.4246, device='cuda:0'), 'train_loss_bbox': tensor(0.1193, device='cuda:0'), 'train_loss_giou': tensor(0.3912, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.1036, device='cuda:0'), 'validation_loss_ce': tensor(0.4565, device='cuda:0'), 'validation_loss_bbox': tensor(0.1409, device='cuda:0'), 'validation_loss_giou': tensor(0.4713, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.4431, device='cuda:0'), 'train_loss_ce': tensor(0.4050, device='cuda:0'), 'train_loss_bbox': tensor(0.1342, device='cuda:0'), 'train_loss_giou': tensor(0.6835, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0913, device='cuda:0'), 'validation_loss_ce': tensor(0.4534, device='cuda:0'), 'validation_loss_bbox': tensor(0.1405, device='cuda:0'), 'validation_loss_giou': tensor(0.4676, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.5937, device='cuda:0'), 'train_loss_ce': tensor(0.4950, device='cuda:0'), 'train_loss_bbox': tensor(0.0820, device='cuda:0'), 'train_loss_giou': tensor(0.3444, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.1049, device='cuda:0'), 'validation_loss_ce': tensor(0.4502, device='cuda:0'), 'validation_loss_bbox': tensor(0.1431, device='cuda:0'), 'validation_loss_giou': tensor(0.4696, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.4078, device='cuda:0'), 'train_loss_ce': tensor(0.3842, device='cuda:0'), 'train_loss_bbox': tensor(0.1092, device='cuda:0'), 'train_loss_giou': tensor(0.2389, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0739, device='cuda:0'), 'validation_loss_ce': tensor(0.4510, device='cuda:0'), 'validation_loss_bbox': tensor(0.1392, device='cuda:0'), 'validation_loss_giou': tensor(0.4634, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.7135, device='cuda:0'), 'train_loss_ce': tensor(0.3938, device='cuda:0'), 'train_loss_bbox': tensor(0.1142, device='cuda:0'), 'train_loss_giou': tensor(0.3742, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0500, device='cuda:0'), 'validation_loss_ce': tensor(0.4449, device='cuda:0'), 'validation_loss_bbox': tensor(0.1333, device='cuda:0'), 'validation_loss_giou': tensor(0.4693, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.8719, device='cuda:0'), 'train_loss_ce': tensor(0.4170, device='cuda:0'), 'train_loss_bbox': tensor(0.0880, device='cuda:0'), 'train_loss_giou': tensor(0.5075, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9737, device='cuda:0'), 'validation_loss_ce': tensor(0.4429, device='cuda:0'), 'validation_loss_bbox': tensor(0.1294, device='cuda:0'), 'validation_loss_giou': tensor(0.4419, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.9734, device='cuda:0'), 'train_loss_ce': tensor(0.4615, device='cuda:0'), 'train_loss_bbox': tensor(0.1003, device='cuda:0'), 'train_loss_giou': tensor(0.5052, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9729, device='cuda:0'), 'validation_loss_ce': tensor(0.4415, device='cuda:0'), 'validation_loss_bbox': tensor(0.1275, device='cuda:0'), 'validation_loss_giou': tensor(0.4470, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.9288, device='cuda:0'), 'train_loss_ce': tensor(0.4898, device='cuda:0'), 'train_loss_bbox': tensor(0.0967, device='cuda:0'), 'train_loss_giou': tensor(0.4778, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0077, device='cuda:0'), 'validation_loss_ce': tensor(0.4397, device='cuda:0'), 'validation_loss_bbox': tensor(0.1310, device='cuda:0'), 'validation_loss_giou': tensor(0.4565, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.5813, device='cuda:0'), 'train_loss_ce': tensor(0.3213, device='cuda:0'), 'train_loss_bbox': tensor(0.0890, device='cuda:0'), 'train_loss_giou': tensor(0.4076, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0185, device='cuda:0'), 'validation_loss_ce': tensor(0.4393, device='cuda:0'), 'validation_loss_bbox': tensor(0.1303, device='cuda:0'), 'validation_loss_giou': tensor(0.4639, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.5830, device='cuda:0'), 'train_loss_ce': tensor(0.3790, device='cuda:0'), 'train_loss_bbox': tensor(0.1060, device='cuda:0'), 'train_loss_giou': tensor(0.3371, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0351, device='cuda:0'), 'validation_loss_ce': tensor(0.4372, device='cuda:0'), 'validation_loss_bbox': tensor(0.1342, device='cuda:0'), 'validation_loss_giou': tensor(0.4634, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2645, device='cuda:0'), 'train_loss_ce': tensor(0.4014, device='cuda:0'), 'train_loss_bbox': tensor(0.0714, device='cuda:0'), 'train_loss_giou': tensor(0.2532, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9948, device='cuda:0'), 'validation_loss_ce': tensor(0.4346, device='cuda:0'), 'validation_loss_bbox': tensor(0.1306, device='cuda:0'), 'validation_loss_giou': tensor(0.4536, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3426, device='cuda:0'), 'train_loss_ce': tensor(0.3139, device='cuda:0'), 'train_loss_bbox': tensor(0.0778, device='cuda:0'), 'train_loss_giou': tensor(0.3199, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9602, device='cuda:0'), 'validation_loss_ce': tensor(0.4345, device='cuda:0'), 'validation_loss_bbox': tensor(0.1249, device='cuda:0'), 'validation_loss_giou': tensor(0.4505, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.5074, device='cuda:0'), 'train_loss_ce': tensor(0.4651, device='cuda:0'), 'train_loss_bbox': tensor(0.0795, device='cuda:0'), 'train_loss_giou': tensor(0.3223, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9882, device='cuda:0'), 'validation_loss_ce': tensor(0.4357, device='cuda:0'), 'validation_loss_bbox': tensor(0.1282, device='cuda:0'), 'validation_loss_giou': tensor(0.4558, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.6619, device='cuda:0'), 'train_loss_ce': tensor(0.4541, device='cuda:0'), 'train_loss_bbox': tensor(0.1077, device='cuda:0'), 'train_loss_giou': tensor(0.3347, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9714, device='cuda:0'), 'validation_loss_ce': tensor(0.4288, device='cuda:0'), 'validation_loss_bbox': tensor(0.1264, device='cuda:0'), 'validation_loss_giou': tensor(0.4554, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2751, device='cuda:0'), 'train_loss_ce': tensor(0.4780, device='cuda:0'), 'train_loss_bbox': tensor(0.0599, device='cuda:0'), 'train_loss_giou': tensor(0.2488, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9359, device='cuda:0'), 'validation_loss_ce': tensor(0.4287, device='cuda:0'), 'validation_loss_bbox': tensor(0.1222, device='cuda:0'), 'validation_loss_giou': tensor(0.4481, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3953, device='cuda:0'), 'train_loss_ce': tensor(0.4210, device='cuda:0'), 'train_loss_bbox': tensor(0.0698, device='cuda:0'), 'train_loss_giou': tensor(0.3127, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9527, device='cuda:0'), 'validation_loss_ce': tensor(0.4282, device='cuda:0'), 'validation_loss_bbox': tensor(0.1249, device='cuda:0'), 'validation_loss_giou': tensor(0.4501, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2062, device='cuda:0'), 'train_loss_ce': tensor(0.4534, device='cuda:0'), 'train_loss_bbox': tensor(0.0659, device='cuda:0'), 'train_loss_giou': tensor(0.2117, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9332, device='cuda:0'), 'validation_loss_ce': tensor(0.4249, device='cuda:0'), 'validation_loss_bbox': tensor(0.1223, device='cuda:0'), 'validation_loss_giou': tensor(0.4484, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.5195, device='cuda:0'), 'train_loss_ce': tensor(0.3928, device='cuda:0'), 'train_loss_bbox': tensor(0.1094, device='cuda:0'), 'train_loss_giou': tensor(0.2898, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9284, device='cuda:0'), 'validation_loss_ce': tensor(0.4224, device='cuda:0'), 'validation_loss_bbox': tensor(0.1230, device='cuda:0'), 'validation_loss_giou': tensor(0.4454, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.1695, device='cuda:0'), 'train_loss_ce': tensor(0.3714, device='cuda:0'), 'train_loss_bbox': tensor(0.0564, device='cuda:0'), 'train_loss_giou': tensor(0.2579, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9271, device='cuda:0'), 'validation_loss_ce': tensor(0.4234, device='cuda:0'), 'validation_loss_bbox': tensor(0.1215, device='cuda:0'), 'validation_loss_giou': tensor(0.4481, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.4215, device='cuda:0'), 'train_loss_ce': tensor(0.4881, device='cuda:0'), 'train_loss_bbox': tensor(0.0761, device='cuda:0'), 'train_loss_giou': tensor(0.2763, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9468, device='cuda:0'), 'validation_loss_ce': tensor(0.4214, device='cuda:0'), 'validation_loss_bbox': tensor(0.1246, device='cuda:0'), 'validation_loss_giou': tensor(0.4512, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3969, device='cuda:0'), 'train_loss_ce': tensor(0.4695, device='cuda:0'), 'train_loss_bbox': tensor(0.0575, device='cuda:0'), 'train_loss_giou': tensor(0.3200, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9221, device='cuda:0'), 'validation_loss_ce': tensor(0.4188, device='cuda:0'), 'validation_loss_bbox': tensor(0.1234, device='cuda:0'), 'validation_loss_giou': tensor(0.4433, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3017, device='cuda:0'), 'train_loss_ce': tensor(0.3737, device='cuda:0'), 'train_loss_bbox': tensor(0.0451, device='cuda:0'), 'train_loss_giou': tensor(0.3513, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9424, device='cuda:0'), 'validation_loss_ce': tensor(0.4178, device='cuda:0'), 'validation_loss_bbox': tensor(0.1247, device='cuda:0'), 'validation_loss_giou': tensor(0.4505, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8226, device='cuda:0'), 'train_loss_ce': tensor(0.2608, device='cuda:0'), 'train_loss_bbox': tensor(0.0339, device='cuda:0'), 'train_loss_giou': tensor(0.1962, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9638, device='cuda:0'), 'validation_loss_ce': tensor(0.4177, device='cuda:0'), 'validation_loss_bbox': tensor(0.1261, device='cuda:0'), 'validation_loss_giou': tensor(0.4579, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9972, device='cuda:0'), 'train_loss_ce': tensor(0.4199, device='cuda:0'), 'train_loss_bbox': tensor(0.0402, device='cuda:0'), 'train_loss_giou': tensor(0.1882, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9488, device='cuda:0'), 'validation_loss_ce': tensor(0.4148, device='cuda:0'), 'validation_loss_bbox': tensor(0.1252, device='cuda:0'), 'validation_loss_giou': tensor(0.4539, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2127, device='cuda:0'), 'train_loss_ce': tensor(0.3963, device='cuda:0'), 'train_loss_bbox': tensor(0.0675, device='cuda:0'), 'train_loss_giou': tensor(0.2394, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9382, device='cuda:0'), 'validation_loss_ce': tensor(0.4113, device='cuda:0'), 'validation_loss_bbox': tensor(0.1234, device='cuda:0'), 'validation_loss_giou': tensor(0.4550, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.4711, device='cuda:0'), 'train_loss_ce': tensor(0.4774, device='cuda:0'), 'train_loss_bbox': tensor(0.0708, device='cuda:0'), 'train_loss_giou': tensor(0.3198, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9266, device='cuda:0'), 'validation_loss_ce': tensor(0.4101, device='cuda:0'), 'validation_loss_bbox': tensor(0.1233, device='cuda:0'), 'validation_loss_giou': tensor(0.4500, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9959, device='cuda:0'), 'train_loss_ce': tensor(0.3633, device='cuda:0'), 'train_loss_bbox': tensor(0.0561, device='cuda:0'), 'train_loss_giou': tensor(0.1760, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9331, device='cuda:0'), 'validation_loss_ce': tensor(0.4114, device='cuda:0'), 'validation_loss_bbox': tensor(0.1232, device='cuda:0'), 'validation_loss_giou': tensor(0.4529, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2761, device='cuda:0'), 'train_loss_ce': tensor(0.3369, device='cuda:0'), 'train_loss_bbox': tensor(0.0609, device='cuda:0'), 'train_loss_giou': tensor(0.3174, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8959, device='cuda:0'), 'validation_loss_ce': tensor(0.4092, device='cuda:0'), 'validation_loss_bbox': tensor(0.1196, device='cuda:0'), 'validation_loss_giou': tensor(0.4443, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.1052, device='cuda:0'), 'train_loss_ce': tensor(0.4021, device='cuda:0'), 'train_loss_bbox': tensor(0.0450, device='cuda:0'), 'train_loss_giou': tensor(0.2390, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9103, device='cuda:0'), 'validation_loss_ce': tensor(0.4082, device='cuda:0'), 'validation_loss_bbox': tensor(0.1223, device='cuda:0'), 'validation_loss_giou': tensor(0.4453, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0489, device='cuda:0'), 'train_loss_ce': tensor(0.4296, device='cuda:0'), 'train_loss_bbox': tensor(0.0439, device='cuda:0'), 'train_loss_giou': tensor(0.2000, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9410, device='cuda:0'), 'validation_loss_ce': tensor(0.4067, device='cuda:0'), 'validation_loss_bbox': tensor(0.1275, device='cuda:0'), 'validation_loss_giou': tensor(0.4484, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0332, device='cuda:0'), 'train_loss_ce': tensor(0.3963, device='cuda:0'), 'train_loss_bbox': tensor(0.0597, device='cuda:0'), 'train_loss_giou': tensor(0.1693, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9021, device='cuda:0'), 'validation_loss_ce': tensor(0.4044, device='cuda:0'), 'validation_loss_bbox': tensor(0.1204, device='cuda:0'), 'validation_loss_giou': tensor(0.4479, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3650, device='cuda:0'), 'train_loss_ce': tensor(0.4742, device='cuda:0'), 'train_loss_bbox': tensor(0.0558, device='cuda:0'), 'train_loss_giou': tensor(0.3059, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9119, device='cuda:0'), 'validation_loss_ce': tensor(0.4054, device='cuda:0'), 'validation_loss_bbox': tensor(0.1221, device='cuda:0'), 'validation_loss_giou': tensor(0.4479, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2805, device='cuda:0'), 'train_loss_ce': tensor(0.4037, device='cuda:0'), 'train_loss_bbox': tensor(0.0486, device='cuda:0'), 'train_loss_giou': tensor(0.3169, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8817, device='cuda:0'), 'validation_loss_ce': tensor(0.4047, device='cuda:0'), 'validation_loss_bbox': tensor(0.1203, device='cuda:0'), 'validation_loss_giou': tensor(0.4377, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.1888, device='cuda:0'), 'train_loss_ce': tensor(0.4283, device='cuda:0'), 'train_loss_bbox': tensor(0.0574, device='cuda:0'), 'train_loss_giou': tensor(0.2369, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8818, device='cuda:0'), 'validation_loss_ce': tensor(0.4025, device='cuda:0'), 'validation_loss_bbox': tensor(0.1179, device='cuda:0'), 'validation_loss_giou': tensor(0.4451, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7404, device='cuda:0'), 'train_loss_ce': tensor(0.3634, device='cuda:0'), 'train_loss_bbox': tensor(0.0316, device='cuda:0'), 'train_loss_giou': tensor(0.1096, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8977, device='cuda:0'), 'validation_loss_ce': tensor(0.4013, device='cuda:0'), 'validation_loss_bbox': tensor(0.1235, device='cuda:0'), 'validation_loss_giou': tensor(0.4395, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0661, device='cuda:0'), 'train_loss_ce': tensor(0.4053, device='cuda:0'), 'train_loss_bbox': tensor(0.0441, device='cuda:0'), 'train_loss_giou': tensor(0.2203, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8821, device='cuda:0'), 'validation_loss_ce': tensor(0.3989, device='cuda:0'), 'validation_loss_bbox': tensor(0.1186, device='cuda:0'), 'validation_loss_giou': tensor(0.4450, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7028, device='cuda:0'), 'train_loss_ce': tensor(0.3904, device='cuda:0'), 'train_loss_bbox': tensor(0.0235, device='cuda:0'), 'train_loss_giou': tensor(0.0974, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.9134, device='cuda:0'), 'validation_loss_ce': tensor(0.3996, device='cuda:0'), 'validation_loss_bbox': tensor(0.1234, device='cuda:0'), 'validation_loss_giou': tensor(0.4485, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7978, device='cuda:0'), 'train_loss_ce': tensor(0.3985, device='cuda:0'), 'train_loss_bbox': tensor(0.0377, device='cuda:0'), 'train_loss_giou': tensor(0.1054, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9087, device='cuda:0'), 'validation_loss_ce': tensor(0.3957, device='cuda:0'), 'validation_loss_bbox': tensor(0.1230, device='cuda:0'), 'validation_loss_giou': tensor(0.4489, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9359, device='cuda:0'), 'train_loss_ce': tensor(0.4104, device='cuda:0'), 'train_loss_bbox': tensor(0.0329, device='cuda:0'), 'train_loss_giou': tensor(0.1804, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8785, device='cuda:0'), 'validation_loss_ce': tensor(0.3954, device='cuda:0'), 'validation_loss_bbox': tensor(0.1187, device='cuda:0'), 'validation_loss_giou': tensor(0.4448, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.4489, device='cuda:0'), 'train_loss_ce': tensor(0.3335, device='cuda:0'), 'train_loss_bbox': tensor(0.0751, device='cuda:0'), 'train_loss_giou': tensor(0.3699, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8740, device='cuda:0'), 'validation_loss_ce': tensor(0.3949, device='cuda:0'), 'validation_loss_bbox': tensor(0.1199, device='cuda:0'), 'validation_loss_giou': tensor(0.4397, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8472, device='cuda:0'), 'train_loss_ce': tensor(0.3181, device='cuda:0'), 'train_loss_bbox': tensor(0.0378, device='cuda:0'), 'train_loss_giou': tensor(0.1699, device='cuda:0'), 'train_cardinality_error': tensor(0.8000, device='cuda:0'), 'validation_loss': tensor(1.8785, device='cuda:0'), 'validation_loss_ce': tensor(0.3940, device='cuda:0'), 'validation_loss_bbox': tensor(0.1190, device='cuda:0'), 'validation_loss_giou': tensor(0.4448, device='cuda:0'), 'validation_cardinality_error': tensor(0.9697, device='cuda:0')}
+{'training_loss': tensor(1.0123, device='cuda:0'), 'train_loss_ce': tensor(0.4111, device='cuda:0'), 'train_loss_bbox': tensor(0.0362, device='cuda:0'), 'train_loss_giou': tensor(0.2100, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9187, device='cuda:0'), 'validation_loss_ce': tensor(0.3924, device='cuda:0'), 'validation_loss_bbox': tensor(0.1253, device='cuda:0'), 'validation_loss_giou': tensor(0.4500, device='cuda:0'), 'validation_cardinality_error': tensor(0.9899, device='cuda:0')}
+{'training_loss': tensor(0.5481, device='cuda:0'), 'train_loss_ce': tensor(0.2719, device='cuda:0'), 'train_loss_bbox': tensor(0.0257, device='cuda:0'), 'train_loss_giou': tensor(0.0739, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9145, device='cuda:0'), 'validation_loss_ce': tensor(0.3937, device='cuda:0'), 'validation_loss_bbox': tensor(0.1219, device='cuda:0'), 'validation_loss_giou': tensor(0.4556, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0651, device='cuda:0'), 'train_loss_ce': tensor(0.3829, device='cuda:0'), 'train_loss_bbox': tensor(0.0515, device='cuda:0'), 'train_loss_giou': tensor(0.2123, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8978, device='cuda:0'), 'validation_loss_ce': tensor(0.3919, device='cuda:0'), 'validation_loss_bbox': tensor(0.1215, device='cuda:0'), 'validation_loss_giou': tensor(0.4491, device='cuda:0'), 'validation_cardinality_error': tensor(0.9899, device='cuda:0')}
 ```
 
 ## Examples