diff --git "a/README.md" "b/README.md"
--- "a/README.md"
+++ "b/README.md"
@@ -3,197 +3,350 @@ library_name: transformers
 tags: []
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
-
-## Model Details
-
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-
-### Model Sources [optional]
-
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-
-## Uses
-
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
-
-## Bias, Risks, and Limitations
-
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
-
-### Recommendations
-
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
-
-## Evaluation
-
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
-
-#### Metrics
-
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
-
-### Results
-
-[More Information Needed]
-
-#### Summary
-
-
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-
-## Technical Specifications [optional]
-
-### Model Architecture and Objective
-
-[More Information Needed]
-
-### Compute Infrastructure
-
-[More Information Needed]
-
-#### Hardware
-
-[More Information Needed]
-
-#### Software
-
-[More Information Needed]
-
-## Citation [optional]
-
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-
-**BibTeX:**
-
-[More Information Needed]
-
-**APA:**
-
-[More Information Needed]
-
-## Glossary [optional]
-
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-
-[More Information Needed]
-
-## More Information [optional]
-
-[More Information Needed]
-
-## Model Card Authors [optional]
-
-[More Information Needed]
-
-## Model Card Contact
-
-[More Information Needed]
\ No newline at end of file
+## 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 ] = 0.000
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.000
+ 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
+```
+
+## After training result
+```
+IoU metric: bbox
+ Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.670
+ Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.800
+ Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.700
+ 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 ] = 0.450
+ Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.725
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.600
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.740
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.740
+ 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 ] = 0.800
+ Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.725
+```
+
+## Config
+- dataset: NIH
+- original model: facebook/detr-resnet-50
+- lr: 0.0001
+- dropout_rate: 0.1
+- weight_decay: 0.0001
+- max_epochs: 300
+
+## Logging
+### Training process
+```
+{'validation_loss': tensor(5.2702, device='cuda:0'), 'validation_loss_ce': tensor(2.1537, device='cuda:0'), 'validation_loss_bbox': tensor(0.3730, device='cuda:0'), 'validation_loss_giou': tensor(0.6258, device='cuda:0'), 'validation_cardinality_error': tensor(98., device='cuda:0')}
+{'training_loss': tensor(5.2702, device='cuda:0'), 'train_loss_ce': tensor(2.1537, device='cuda:0'), 'train_loss_bbox': tensor(0.3730, device='cuda:0'), 'train_loss_giou': tensor(0.6258, device='cuda:0'), 'train_cardinality_error': tensor(98., device='cuda:0'), 'validation_loss': tensor(6.0122, device='cuda:0'), 'validation_loss_ce': tensor(2.1009, device='cuda:0'), 'validation_loss_bbox': tensor(0.4820, device='cuda:0'), 'validation_loss_giou': tensor(0.7507, device='cuda:0'), 'validation_cardinality_error': tensor(95.8000, device='cuda:0')}
+{'training_loss': tensor(6.0122, device='cuda:0'), 'train_loss_ce': tensor(2.1009, device='cuda:0'), 'train_loss_bbox': tensor(0.4820, device='cuda:0'), 'train_loss_giou': tensor(0.7507, device='cuda:0'), 'train_cardinality_error': tensor(95.8000, device='cuda:0'), 'validation_loss': tensor(5.8762, device='cuda:0'), 'validation_loss_ce': tensor(1.9713, device='cuda:0'), 'validation_loss_bbox': tensor(0.4859, device='cuda:0'), 'validation_loss_giou': tensor(0.7377, device='cuda:0'), 'validation_cardinality_error': tensor(75.2000, device='cuda:0')}
+{'training_loss': tensor(5.8762, device='cuda:0'), 'train_loss_ce': tensor(1.9713, device='cuda:0'), 'train_loss_bbox': tensor(0.4859, device='cuda:0'), 'train_loss_giou': tensor(0.7377, device='cuda:0'), 'train_cardinality_error': tensor(75.2000, device='cuda:0'), 'validation_loss': tensor(4.0028, device='cuda:0'), 'validation_loss_ce': tensor(1.9185, device='cuda:0'), 'validation_loss_bbox': tensor(0.1857, device='cuda:0'), 'validation_loss_giou': tensor(0.5778, device='cuda:0'), 'validation_cardinality_error': tensor(53., device='cuda:0')}
+{'training_loss': tensor(4.0028, device='cuda:0'), 'train_loss_ce': tensor(1.9185, device='cuda:0'), 'train_loss_bbox': tensor(0.1857, device='cuda:0'), 'train_loss_giou': tensor(0.5778, device='cuda:0'), 'train_cardinality_error': tensor(53., device='cuda:0'), 'validation_loss': tensor(4.1011, device='cuda:0'), 'validation_loss_ce': tensor(1.8516, device='cuda:0'), 'validation_loss_bbox': tensor(0.2148, device='cuda:0'), 'validation_loss_giou': tensor(0.5876, device='cuda:0'), 'validation_cardinality_error': tensor(23.4000, device='cuda:0')}
+{'training_loss': tensor(4.1011, device='cuda:0'), 'train_loss_ce': tensor(1.8516, device='cuda:0'), 'train_loss_bbox': tensor(0.2148, device='cuda:0'), 'train_loss_giou': tensor(0.5876, device='cuda:0'), 'train_cardinality_error': tensor(23.4000, device='cuda:0'), 'validation_loss': tensor(3.8432, device='cuda:0'), 'validation_loss_ce': tensor(1.7577, device='cuda:0'), 'validation_loss_bbox': tensor(0.1873, device='cuda:0'), 'validation_loss_giou': tensor(0.5744, device='cuda:0'), 'validation_cardinality_error': tensor(5., device='cuda:0')}
+{'training_loss': tensor(3.8432, device='cuda:0'), 'train_loss_ce': tensor(1.7577, device='cuda:0'), 'train_loss_bbox': tensor(0.1873, device='cuda:0'), 'train_loss_giou': tensor(0.5744, device='cuda:0'), 'train_cardinality_error': tensor(5., device='cuda:0'), 'validation_loss': tensor(3.1571, device='cuda:0'), 'validation_loss_ce': tensor(1.7022, device='cuda:0'), 'validation_loss_bbox': tensor(0.1197, device='cuda:0'), 'validation_loss_giou': tensor(0.4281, device='cuda:0'), 'validation_cardinality_error': tensor(2.6000, device='cuda:0')}
+{'training_loss': tensor(3.1571, device='cuda:0'), 'train_loss_ce': tensor(1.7022, device='cuda:0'), 'train_loss_bbox': tensor(0.1197, device='cuda:0'), 'train_loss_giou': tensor(0.4281, device='cuda:0'), 'train_cardinality_error': tensor(2.6000, device='cuda:0'), 'validation_loss': tensor(4.1719, device='cuda:0'), 'validation_loss_ce': tensor(1.6694, device='cuda:0'), 'validation_loss_bbox': tensor(0.2118, device='cuda:0'), 'validation_loss_giou': tensor(0.7216, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(4.1719, device='cuda:0'), 'train_loss_ce': tensor(1.6694, device='cuda:0'), 'train_loss_bbox': tensor(0.2118, device='cuda:0'), 'train_loss_giou': tensor(0.7216, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.8560, device='cuda:0'), 'validation_loss_ce': tensor(1.5656, device='cuda:0'), 'validation_loss_bbox': tensor(0.2503, device='cuda:0'), 'validation_loss_giou': tensor(0.5194, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.8560, device='cuda:0'), 'train_loss_ce': tensor(1.5656, device='cuda:0'), 'train_loss_bbox': tensor(0.2503, device='cuda:0'), 'train_loss_giou': tensor(0.5194, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.4982, device='cuda:0'), 'validation_loss_ce': tensor(1.5151, device='cuda:0'), 'validation_loss_bbox': tensor(0.1873, device='cuda:0'), 'validation_loss_giou': tensor(0.5232, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.4982, device='cuda:0'), 'train_loss_ce': tensor(1.5151, device='cuda:0'), 'train_loss_bbox': tensor(0.1873, device='cuda:0'), 'train_loss_giou': tensor(0.5232, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.6987, device='cuda:0'), 'validation_loss_ce': tensor(1.4569, device='cuda:0'), 'validation_loss_bbox': tensor(0.2359, device='cuda:0'), 'validation_loss_giou': tensor(0.5311, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.6987, device='cuda:0'), 'train_loss_ce': tensor(1.4569, device='cuda:0'), 'train_loss_bbox': tensor(0.2359, device='cuda:0'), 'train_loss_giou': tensor(0.5311, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(5.4486, device='cuda:0'), 'validation_loss_ce': tensor(1.3981, device='cuda:0'), 'validation_loss_bbox': tensor(0.5005, device='cuda:0'), 'validation_loss_giou': tensor(0.7739, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(5.4486, device='cuda:0'), 'train_loss_ce': tensor(1.3981, device='cuda:0'), 'train_loss_bbox': tensor(0.5005, device='cuda:0'), 'train_loss_giou': tensor(0.7739, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(4.3868, device='cuda:0'), 'validation_loss_ce': tensor(1.3238, device='cuda:0'), 'validation_loss_bbox': tensor(0.3483, device='cuda:0'), 'validation_loss_giou': tensor(0.6609, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(4.3868, device='cuda:0'), 'train_loss_ce': tensor(1.3238, device='cuda:0'), 'train_loss_bbox': tensor(0.3483, device='cuda:0'), 'train_loss_giou': tensor(0.6609, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.8680, device='cuda:0'), 'validation_loss_ce': tensor(1.2573, device='cuda:0'), 'validation_loss_bbox': tensor(0.2932, device='cuda:0'), 'validation_loss_giou': tensor(0.5724, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.8680, device='cuda:0'), 'train_loss_ce': tensor(1.2573, device='cuda:0'), 'train_loss_bbox': tensor(0.2932, device='cuda:0'), 'train_loss_giou': tensor(0.5724, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.4306, device='cuda:0'), 'validation_loss_ce': tensor(1.1976, device='cuda:0'), 'validation_loss_bbox': tensor(0.2274, device='cuda:0'), 'validation_loss_giou': tensor(0.5481, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.4306, device='cuda:0'), 'train_loss_ce': tensor(1.1976, device='cuda:0'), 'train_loss_bbox': tensor(0.2274, device='cuda:0'), 'train_loss_giou': tensor(0.5481, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.7234, device='cuda:0'), 'validation_loss_ce': tensor(1.1880, device='cuda:0'), 'validation_loss_bbox': tensor(0.2853, device='cuda:0'), 'validation_loss_giou': tensor(0.5545, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.7234, device='cuda:0'), 'train_loss_ce': tensor(1.1880, device='cuda:0'), 'train_loss_bbox': tensor(0.2853, device='cuda:0'), 'train_loss_giou': tensor(0.5545, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.9480, device='cuda:0'), 'validation_loss_ce': tensor(1.1779, device='cuda:0'), 'validation_loss_bbox': tensor(0.2935, device='cuda:0'), 'validation_loss_giou': tensor(0.6512, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.9480, device='cuda:0'), 'train_loss_ce': tensor(1.1779, device='cuda:0'), 'train_loss_bbox': tensor(0.2935, device='cuda:0'), 'train_loss_giou': tensor(0.6512, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.2322, device='cuda:0'), 'validation_loss_ce': tensor(1.1211, device='cuda:0'), 'validation_loss_bbox': tensor(0.1940, device='cuda:0'), 'validation_loss_giou': tensor(0.5705, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.2322, device='cuda:0'), 'train_loss_ce': tensor(1.1211, device='cuda:0'), 'train_loss_bbox': tensor(0.1940, device='cuda:0'), 'train_loss_giou': tensor(0.5705, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(3.0639, device='cuda:0'), 'validation_loss_ce': tensor(1.0811, device='cuda:0'), 'validation_loss_bbox': tensor(0.1720, device='cuda:0'), 'validation_loss_giou': tensor(0.5613, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(3.0639, device='cuda:0'), 'train_loss_ce': tensor(1.0811, device='cuda:0'), 'train_loss_bbox': tensor(0.1720, device='cuda:0'), 'train_loss_giou': tensor(0.5613, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.7318, device='cuda:0'), 'validation_loss_ce': tensor(1.0376, device='cuda:0'), 'validation_loss_bbox': tensor(0.1484, device='cuda:0'), 'validation_loss_giou': tensor(0.4761, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.7318, device='cuda:0'), 'train_loss_ce': tensor(1.0376, device='cuda:0'), 'train_loss_bbox': tensor(0.1484, device='cuda:0'), 'train_loss_giou': tensor(0.4761, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.6254, device='cuda:0'), 'validation_loss_ce': tensor(0.9964, device='cuda:0'), 'validation_loss_bbox': tensor(0.1325, device='cuda:0'), 'validation_loss_giou': tensor(0.4833, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.6254, device='cuda:0'), 'train_loss_ce': tensor(0.9964, device='cuda:0'), 'train_loss_bbox': tensor(0.1325, device='cuda:0'), 'train_loss_giou': tensor(0.4833, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.4947, device='cuda:0'), 'validation_loss_ce': tensor(0.9665, device='cuda:0'), 'validation_loss_bbox': tensor(0.1258, device='cuda:0'), 'validation_loss_giou': tensor(0.4495, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.4947, device='cuda:0'), 'train_loss_ce': tensor(0.9665, device='cuda:0'), 'train_loss_bbox': tensor(0.1258, device='cuda:0'), 'train_loss_giou': tensor(0.4495, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2255, device='cuda:0'), 'validation_loss_ce': tensor(0.9455, device='cuda:0'), 'validation_loss_bbox': tensor(0.0844, device='cuda:0'), 'validation_loss_giou': tensor(0.4290, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.2255, device='cuda:0'), 'train_loss_ce': tensor(0.9455, device='cuda:0'), 'train_loss_bbox': tensor(0.0844, device='cuda:0'), 'train_loss_giou': tensor(0.4290, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.3806, device='cuda:0'), 'validation_loss_ce': tensor(0.9262, device='cuda:0'), 'validation_loss_bbox': tensor(0.1182, device='cuda:0'), 'validation_loss_giou': tensor(0.4317, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.3806, device='cuda:0'), 'train_loss_ce': tensor(0.9262, device='cuda:0'), 'train_loss_bbox': tensor(0.1182, device='cuda:0'), 'train_loss_giou': tensor(0.4317, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8096, device='cuda:0'), 'validation_loss_ce': tensor(0.9066, device='cuda:0'), 'validation_loss_bbox': tensor(0.0652, device='cuda:0'), 'validation_loss_giou': tensor(0.2884, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.8096, device='cuda:0'), 'train_loss_ce': tensor(0.9066, device='cuda:0'), 'train_loss_bbox': tensor(0.0652, device='cuda:0'), 'train_loss_giou': tensor(0.2884, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.1115, device='cuda:0'), 'validation_loss_ce': tensor(0.8959, device='cuda:0'), 'validation_loss_bbox': tensor(0.1066, device='cuda:0'), 'validation_loss_giou': tensor(0.3412, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.1115, device='cuda:0'), 'train_loss_ce': tensor(0.8959, device='cuda:0'), 'train_loss_bbox': tensor(0.1066, device='cuda:0'), 'train_loss_giou': tensor(0.3412, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.3829, device='cuda:0'), 'validation_loss_ce': tensor(0.8773, device='cuda:0'), 'validation_loss_bbox': tensor(0.1257, device='cuda:0'), 'validation_loss_giou': tensor(0.4386, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.3829, device='cuda:0'), 'train_loss_ce': tensor(0.8773, device='cuda:0'), 'train_loss_bbox': tensor(0.1257, device='cuda:0'), 'train_loss_giou': tensor(0.4386, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.3398, device='cuda:0'), 'validation_loss_ce': tensor(0.8451, device='cuda:0'), 'validation_loss_bbox': tensor(0.1202, device='cuda:0'), 'validation_loss_giou': tensor(0.4468, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.3398, device='cuda:0'), 'train_loss_ce': tensor(0.8451, device='cuda:0'), 'train_loss_bbox': tensor(0.1202, device='cuda:0'), 'train_loss_giou': tensor(0.4468, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.0513, device='cuda:0'), 'validation_loss_ce': tensor(0.7986, device='cuda:0'), 'validation_loss_bbox': tensor(0.1040, device='cuda:0'), 'validation_loss_giou': tensor(0.3663, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.0513, device='cuda:0'), 'train_loss_ce': tensor(0.7986, device='cuda:0'), 'train_loss_bbox': tensor(0.1040, device='cuda:0'), 'train_loss_giou': tensor(0.3663, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(2.2376, device='cuda:0'), 'validation_loss_ce': tensor(0.7650, device='cuda:0'), 'validation_loss_bbox': tensor(0.1324, device='cuda:0'), 'validation_loss_giou': tensor(0.4053, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(2.2376, device='cuda:0'), 'train_loss_ce': tensor(0.7650, device='cuda:0'), 'train_loss_bbox': tensor(0.1324, device='cuda:0'), 'train_loss_giou': tensor(0.4053, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9457, device='cuda:0'), 'validation_loss_ce': tensor(0.7422, device='cuda:0'), 'validation_loss_bbox': tensor(0.0967, device='cuda:0'), 'validation_loss_giou': tensor(0.3600, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.9457, device='cuda:0'), 'train_loss_ce': tensor(0.7422, device='cuda:0'), 'train_loss_bbox': tensor(0.0967, device='cuda:0'), 'train_loss_giou': tensor(0.3600, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.9051, device='cuda:0'), 'validation_loss_ce': tensor(0.7205, device='cuda:0'), 'validation_loss_bbox': tensor(0.0957, device='cuda:0'), 'validation_loss_giou': tensor(0.3530, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.9051, device='cuda:0'), 'train_loss_ce': tensor(0.7205, device='cuda:0'), 'train_loss_bbox': tensor(0.0957, device='cuda:0'), 'train_loss_giou': tensor(0.3530, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.8620, device='cuda:0'), 'validation_loss_ce': tensor(0.7092, device='cuda:0'), 'validation_loss_bbox': tensor(0.0976, device='cuda:0'), 'validation_loss_giou': tensor(0.3325, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.8620, device='cuda:0'), 'train_loss_ce': tensor(0.7092, device='cuda:0'), 'train_loss_bbox': tensor(0.0976, device='cuda:0'), 'train_loss_giou': tensor(0.3325, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.7245, device='cuda:0'), 'validation_loss_ce': tensor(0.6973, device='cuda:0'), 'validation_loss_bbox': tensor(0.0835, device='cuda:0'), 'validation_loss_giou': tensor(0.3049, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.7245, device='cuda:0'), 'train_loss_ce': tensor(0.6973, device='cuda:0'), 'train_loss_bbox': tensor(0.0835, device='cuda:0'), 'train_loss_giou': tensor(0.3049, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.5914, device='cuda:0'), 'validation_loss_ce': tensor(0.6855, device='cuda:0'), 'validation_loss_bbox': tensor(0.0665, device='cuda:0'), 'validation_loss_giou': tensor(0.2867, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.5914, device='cuda:0'), 'train_loss_ce': tensor(0.6855, device='cuda:0'), 'train_loss_bbox': tensor(0.0665, device='cuda:0'), 'train_loss_giou': tensor(0.2867, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.3430, device='cuda:0'), 'validation_loss_ce': tensor(0.6724, device='cuda:0'), 'validation_loss_bbox': tensor(0.0488, device='cuda:0'), 'validation_loss_giou': tensor(0.2133, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3430, device='cuda:0'), 'train_loss_ce': tensor(0.6724, device='cuda:0'), 'train_loss_bbox': tensor(0.0488, device='cuda:0'), 'train_loss_giou': tensor(0.2133, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.4321, device='cuda:0'), 'validation_loss_ce': tensor(0.6572, device='cuda:0'), 'validation_loss_bbox': tensor(0.0555, device='cuda:0'), 'validation_loss_giou': tensor(0.2488, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.4321, device='cuda:0'), 'train_loss_ce': tensor(0.6572, device='cuda:0'), 'train_loss_bbox': tensor(0.0555, device='cuda:0'), 'train_loss_giou': tensor(0.2488, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.3661, device='cuda:0'), 'validation_loss_ce': tensor(0.6361, device='cuda:0'), 'validation_loss_bbox': tensor(0.0519, device='cuda:0'), 'validation_loss_giou': tensor(0.2353, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.3661, device='cuda:0'), 'train_loss_ce': tensor(0.6361, device='cuda:0'), 'train_loss_bbox': tensor(0.0519, device='cuda:0'), 'train_loss_giou': tensor(0.2353, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.2293, device='cuda:0'), 'validation_loss_ce': tensor(0.6169, device='cuda:0'), 'validation_loss_bbox': tensor(0.0466, device='cuda:0'), 'validation_loss_giou': tensor(0.1897, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2293, device='cuda:0'), 'train_loss_ce': tensor(0.6169, device='cuda:0'), 'train_loss_bbox': tensor(0.0466, device='cuda:0'), 'train_loss_giou': tensor(0.1897, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.2547, device='cuda:0'), 'validation_loss_ce': tensor(0.5942, device='cuda:0'), 'validation_loss_bbox': tensor(0.0489, device='cuda:0'), 'validation_loss_giou': tensor(0.2080, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2547, device='cuda:0'), 'train_loss_ce': tensor(0.5942, device='cuda:0'), 'train_loss_bbox': tensor(0.0489, device='cuda:0'), 'train_loss_giou': tensor(0.2080, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.1843, device='cuda:0'), 'validation_loss_ce': tensor(0.5740, device='cuda:0'), 'validation_loss_bbox': tensor(0.0446, device='cuda:0'), 'validation_loss_giou': tensor(0.1936, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.1843, device='cuda:0'), 'train_loss_ce': tensor(0.5740, device='cuda:0'), 'train_loss_bbox': tensor(0.0446, device='cuda:0'), 'train_loss_giou': tensor(0.1936, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.2412, device='cuda:0'), 'validation_loss_ce': tensor(0.5584, device='cuda:0'), 'validation_loss_bbox': tensor(0.0522, device='cuda:0'), 'validation_loss_giou': tensor(0.2109, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2412, device='cuda:0'), 'train_loss_ce': tensor(0.5584, device='cuda:0'), 'train_loss_bbox': tensor(0.0522, device='cuda:0'), 'train_loss_giou': tensor(0.2109, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.2400, device='cuda:0'), 'validation_loss_ce': tensor(0.5459, device='cuda:0'), 'validation_loss_bbox': tensor(0.0534, device='cuda:0'), 'validation_loss_giou': tensor(0.2136, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.2400, device='cuda:0'), 'train_loss_ce': tensor(0.5459, device='cuda:0'), 'train_loss_bbox': tensor(0.0534, device='cuda:0'), 'train_loss_giou': tensor(0.2136, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.1269, device='cuda:0'), 'validation_loss_ce': tensor(0.5355, device='cuda:0'), 'validation_loss_bbox': tensor(0.0481, device='cuda:0'), 'validation_loss_giou': tensor(0.1755, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.1269, device='cuda:0'), 'train_loss_ce': tensor(0.5355, device='cuda:0'), 'train_loss_bbox': tensor(0.0481, device='cuda:0'), 'train_loss_giou': tensor(0.1755, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.0539, device='cuda:0'), 'validation_loss_ce': tensor(0.5263, device='cuda:0'), 'validation_loss_bbox': tensor(0.0356, device='cuda:0'), 'validation_loss_giou': tensor(0.1748, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0539, device='cuda:0'), 'train_loss_ce': tensor(0.5263, device='cuda:0'), 'train_loss_bbox': tensor(0.0356, device='cuda:0'), 'train_loss_giou': tensor(0.1748, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9615, device='cuda:0'), 'validation_loss_ce': tensor(0.5183, device='cuda:0'), 'validation_loss_bbox': tensor(0.0268, device='cuda:0'), 'validation_loss_giou': tensor(0.1546, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9615, device='cuda:0'), 'train_loss_ce': tensor(0.5183, device='cuda:0'), 'train_loss_bbox': tensor(0.0268, device='cuda:0'), 'train_loss_giou': tensor(0.1546, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.0075, device='cuda:0'), 'validation_loss_ce': tensor(0.5159, device='cuda:0'), 'validation_loss_bbox': tensor(0.0224, device='cuda:0'), 'validation_loss_giou': tensor(0.1899, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0075, device='cuda:0'), 'train_loss_ce': tensor(0.5159, device='cuda:0'), 'train_loss_bbox': tensor(0.0224, device='cuda:0'), 'train_loss_giou': tensor(0.1899, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(1.0135, device='cuda:0'), 'validation_loss_ce': tensor(0.5050, device='cuda:0'), 'validation_loss_bbox': tensor(0.0362, device='cuda:0'), 'validation_loss_giou': tensor(0.1639, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(1.0135, device='cuda:0'), 'train_loss_ce': tensor(0.5050, device='cuda:0'), 'train_loss_bbox': tensor(0.0362, device='cuda:0'), 'train_loss_giou': tensor(0.1639, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9748, device='cuda:0'), 'validation_loss_ce': tensor(0.4942, device='cuda:0'), 'validation_loss_bbox': tensor(0.0305, device='cuda:0'), 'validation_loss_giou': tensor(0.1641, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9748, device='cuda:0'), 'train_loss_ce': tensor(0.4942, device='cuda:0'), 'train_loss_bbox': tensor(0.0305, device='cuda:0'), 'train_loss_giou': tensor(0.1641, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9321, device='cuda:0'), 'validation_loss_ce': tensor(0.4892, device='cuda:0'), 'validation_loss_bbox': tensor(0.0255, device='cuda:0'), 'validation_loss_giou': tensor(0.1577, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9321, device='cuda:0'), 'train_loss_ce': tensor(0.4892, device='cuda:0'), 'train_loss_bbox': tensor(0.0255, device='cuda:0'), 'train_loss_giou': tensor(0.1577, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7590, device='cuda:0'), 'validation_loss_ce': tensor(0.4857, device='cuda:0'), 'validation_loss_bbox': tensor(0.0195, device='cuda:0'), 'validation_loss_giou': tensor(0.0880, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7590, device='cuda:0'), 'train_loss_ce': tensor(0.4857, device='cuda:0'), 'train_loss_bbox': tensor(0.0195, device='cuda:0'), 'train_loss_giou': tensor(0.0880, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8876, device='cuda:0'), 'validation_loss_ce': tensor(0.4823, device='cuda:0'), 'validation_loss_bbox': tensor(0.0174, device='cuda:0'), 'validation_loss_giou': tensor(0.1591, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8876, device='cuda:0'), 'train_loss_ce': tensor(0.4823, device='cuda:0'), 'train_loss_bbox': tensor(0.0174, device='cuda:0'), 'train_loss_giou': tensor(0.1591, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9084, device='cuda:0'), 'validation_loss_ce': tensor(0.4786, device='cuda:0'), 'validation_loss_bbox': tensor(0.0252, device='cuda:0'), 'validation_loss_giou': tensor(0.1518, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9084, device='cuda:0'), 'train_loss_ce': tensor(0.4786, device='cuda:0'), 'train_loss_bbox': tensor(0.0252, device='cuda:0'), 'train_loss_giou': tensor(0.1518, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8691, device='cuda:0'), 'validation_loss_ce': tensor(0.4751, device='cuda:0'), 'validation_loss_bbox': tensor(0.0221, device='cuda:0'), 'validation_loss_giou': tensor(0.1418, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8691, device='cuda:0'), 'train_loss_ce': tensor(0.4751, device='cuda:0'), 'train_loss_bbox': tensor(0.0221, device='cuda:0'), 'train_loss_giou': tensor(0.1418, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7783, device='cuda:0'), 'validation_loss_ce': tensor(0.4702, device='cuda:0'), 'validation_loss_bbox': tensor(0.0151, device='cuda:0'), 'validation_loss_giou': tensor(0.1162, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7783, device='cuda:0'), 'train_loss_ce': tensor(0.4702, device='cuda:0'), 'train_loss_bbox': tensor(0.0151, device='cuda:0'), 'train_loss_giou': tensor(0.1162, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8734, device='cuda:0'), 'validation_loss_ce': tensor(0.4642, device='cuda:0'), 'validation_loss_bbox': tensor(0.0243, device='cuda:0'), 'validation_loss_giou': tensor(0.1439, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8734, device='cuda:0'), 'train_loss_ce': tensor(0.4642, device='cuda:0'), 'train_loss_bbox': tensor(0.0243, device='cuda:0'), 'train_loss_giou': tensor(0.1439, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7773, device='cuda:0'), 'validation_loss_ce': tensor(0.4586, device='cuda:0'), 'validation_loss_bbox': tensor(0.0240, device='cuda:0'), 'validation_loss_giou': tensor(0.0993, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7773, device='cuda:0'), 'train_loss_ce': tensor(0.4586, device='cuda:0'), 'train_loss_bbox': tensor(0.0240, device='cuda:0'), 'train_loss_giou': tensor(0.0993, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9285, device='cuda:0'), 'validation_loss_ce': tensor(0.4558, device='cuda:0'), 'validation_loss_bbox': tensor(0.0294, device='cuda:0'), 'validation_loss_giou': tensor(0.1629, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9285, device='cuda:0'), 'train_loss_ce': tensor(0.4558, device='cuda:0'), 'train_loss_bbox': tensor(0.0294, device='cuda:0'), 'train_loss_giou': tensor(0.1629, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9313, device='cuda:0'), 'validation_loss_ce': tensor(0.4506, device='cuda:0'), 'validation_loss_bbox': tensor(0.0266, device='cuda:0'), 'validation_loss_giou': tensor(0.1740, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9313, device='cuda:0'), 'train_loss_ce': tensor(0.4506, device='cuda:0'), 'train_loss_bbox': tensor(0.0266, device='cuda:0'), 'train_loss_giou': tensor(0.1740, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.9146, device='cuda:0'), 'validation_loss_ce': tensor(0.4455, device='cuda:0'), 'validation_loss_bbox': tensor(0.0258, device='cuda:0'), 'validation_loss_giou': tensor(0.1699, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.9146, device='cuda:0'), 'train_loss_ce': tensor(0.4455, device='cuda:0'), 'train_loss_bbox': tensor(0.0258, device='cuda:0'), 'train_loss_giou': tensor(0.1699, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8368, device='cuda:0'), 'validation_loss_ce': tensor(0.4407, device='cuda:0'), 'validation_loss_bbox': tensor(0.0235, device='cuda:0'), 'validation_loss_giou': tensor(0.1392, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8368, device='cuda:0'), 'train_loss_ce': tensor(0.4407, device='cuda:0'), 'train_loss_bbox': tensor(0.0235, device='cuda:0'), 'train_loss_giou': tensor(0.1392, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7998, device='cuda:0'), 'validation_loss_ce': tensor(0.4366, device='cuda:0'), 'validation_loss_bbox': tensor(0.0234, device='cuda:0'), 'validation_loss_giou': tensor(0.1232, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7998, device='cuda:0'), 'train_loss_ce': tensor(0.4366, device='cuda:0'), 'train_loss_bbox': tensor(0.0234, device='cuda:0'), 'train_loss_giou': tensor(0.1232, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8089, device='cuda:0'), 'validation_loss_ce': tensor(0.4337, device='cuda:0'), 'validation_loss_bbox': tensor(0.0238, device='cuda:0'), 'validation_loss_giou': tensor(0.1281, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8089, device='cuda:0'), 'train_loss_ce': tensor(0.4337, device='cuda:0'), 'train_loss_bbox': tensor(0.0238, device='cuda:0'), 'train_loss_giou': tensor(0.1281, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7928, device='cuda:0'), 'validation_loss_ce': tensor(0.4312, device='cuda:0'), 'validation_loss_bbox': tensor(0.0188, device='cuda:0'), 'validation_loss_giou': tensor(0.1338, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7928, device='cuda:0'), 'train_loss_ce': tensor(0.4312, device='cuda:0'), 'train_loss_bbox': tensor(0.0188, device='cuda:0'), 'train_loss_giou': tensor(0.1338, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6416, device='cuda:0'), 'validation_loss_ce': tensor(0.4283, device='cuda:0'), 'validation_loss_bbox': tensor(0.0127, device='cuda:0'), 'validation_loss_giou': tensor(0.0748, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6416, device='cuda:0'), 'train_loss_ce': tensor(0.4283, device='cuda:0'), 'train_loss_bbox': tensor(0.0127, device='cuda:0'), 'train_loss_giou': tensor(0.0748, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7243, device='cuda:0'), 'validation_loss_ce': tensor(0.4261, device='cuda:0'), 'validation_loss_bbox': tensor(0.0159, device='cuda:0'), 'validation_loss_giou': tensor(0.1092, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7243, device='cuda:0'), 'train_loss_ce': tensor(0.4261, device='cuda:0'), 'train_loss_bbox': tensor(0.0159, device='cuda:0'), 'train_loss_giou': tensor(0.1092, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7855, device='cuda:0'), 'validation_loss_ce': tensor(0.4239, device='cuda:0'), 'validation_loss_bbox': tensor(0.0176, device='cuda:0'), 'validation_loss_giou': tensor(0.1367, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7855, device='cuda:0'), 'train_loss_ce': tensor(0.4239, device='cuda:0'), 'train_loss_bbox': tensor(0.0176, device='cuda:0'), 'train_loss_giou': tensor(0.1367, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7325, device='cuda:0'), 'validation_loss_ce': tensor(0.4219, device='cuda:0'), 'validation_loss_bbox': tensor(0.0163, device='cuda:0'), 'validation_loss_giou': tensor(0.1147, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7325, device='cuda:0'), 'train_loss_ce': tensor(0.4219, device='cuda:0'), 'train_loss_bbox': tensor(0.0163, device='cuda:0'), 'train_loss_giou': tensor(0.1147, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6533, device='cuda:0'), 'validation_loss_ce': tensor(0.4194, device='cuda:0'), 'validation_loss_bbox': tensor(0.0152, device='cuda:0'), 'validation_loss_giou': tensor(0.0790, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6533, device='cuda:0'), 'train_loss_ce': tensor(0.4194, device='cuda:0'), 'train_loss_bbox': tensor(0.0152, device='cuda:0'), 'train_loss_giou': tensor(0.0790, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7328, device='cuda:0'), 'validation_loss_ce': tensor(0.4145, device='cuda:0'), 'validation_loss_bbox': tensor(0.0172, device='cuda:0'), 'validation_loss_giou': tensor(0.1162, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7328, device='cuda:0'), 'train_loss_ce': tensor(0.4145, device='cuda:0'), 'train_loss_bbox': tensor(0.0172, device='cuda:0'), 'train_loss_giou': tensor(0.1162, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7975, device='cuda:0'), 'validation_loss_ce': tensor(0.4149, device='cuda:0'), 'validation_loss_bbox': tensor(0.0239, device='cuda:0'), 'validation_loss_giou': tensor(0.1314, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7975, device='cuda:0'), 'train_loss_ce': tensor(0.4149, device='cuda:0'), 'train_loss_bbox': tensor(0.0239, device='cuda:0'), 'train_loss_giou': tensor(0.1314, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6533, device='cuda:0'), 'validation_loss_ce': tensor(0.4137, device='cuda:0'), 'validation_loss_bbox': tensor(0.0130, device='cuda:0'), 'validation_loss_giou': tensor(0.0874, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6533, device='cuda:0'), 'train_loss_ce': tensor(0.4137, device='cuda:0'), 'train_loss_bbox': tensor(0.0130, device='cuda:0'), 'train_loss_giou': tensor(0.0874, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7546, device='cuda:0'), 'validation_loss_ce': tensor(0.4129, device='cuda:0'), 'validation_loss_bbox': tensor(0.0213, device='cuda:0'), 'validation_loss_giou': tensor(0.1177, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7546, device='cuda:0'), 'train_loss_ce': tensor(0.4129, device='cuda:0'), 'train_loss_bbox': tensor(0.0213, device='cuda:0'), 'train_loss_giou': tensor(0.1177, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6477, device='cuda:0'), 'validation_loss_ce': tensor(0.4121, device='cuda:0'), 'validation_loss_bbox': tensor(0.0172, device='cuda:0'), 'validation_loss_giou': tensor(0.0748, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6477, device='cuda:0'), 'train_loss_ce': tensor(0.4121, device='cuda:0'), 'train_loss_bbox': tensor(0.0172, device='cuda:0'), 'train_loss_giou': tensor(0.0748, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7925, device='cuda:0'), 'validation_loss_ce': tensor(0.4106, device='cuda:0'), 'validation_loss_bbox': tensor(0.0257, device='cuda:0'), 'validation_loss_giou': tensor(0.1267, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7925, device='cuda:0'), 'train_loss_ce': tensor(0.4106, device='cuda:0'), 'train_loss_bbox': tensor(0.0257, device='cuda:0'), 'train_loss_giou': tensor(0.1267, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.8183, device='cuda:0'), 'validation_loss_ce': tensor(0.4088, device='cuda:0'), 'validation_loss_bbox': tensor(0.0251, device='cuda:0'), 'validation_loss_giou': tensor(0.1420, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.8183, device='cuda:0'), 'train_loss_ce': tensor(0.4088, device='cuda:0'), 'train_loss_bbox': tensor(0.0251, device='cuda:0'), 'train_loss_giou': tensor(0.1420, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7163, device='cuda:0'), 'validation_loss_ce': tensor(0.4066, device='cuda:0'), 'validation_loss_bbox': tensor(0.0171, device='cuda:0'), 'validation_loss_giou': tensor(0.1121, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7163, device='cuda:0'), 'train_loss_ce': tensor(0.4066, device='cuda:0'), 'train_loss_bbox': tensor(0.0171, device='cuda:0'), 'train_loss_giou': tensor(0.1121, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7981, device='cuda:0'), 'validation_loss_ce': tensor(0.4042, device='cuda:0'), 'validation_loss_bbox': tensor(0.0247, device='cuda:0'), 'validation_loss_giou': tensor(0.1352, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7981, device='cuda:0'), 'train_loss_ce': tensor(0.4042, device='cuda:0'), 'train_loss_bbox': tensor(0.0247, device='cuda:0'), 'train_loss_giou': tensor(0.1352, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7671, device='cuda:0'), 'validation_loss_ce': tensor(0.4027, device='cuda:0'), 'validation_loss_bbox': tensor(0.0246, device='cuda:0'), 'validation_loss_giou': tensor(0.1206, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7671, device='cuda:0'), 'train_loss_ce': tensor(0.4027, device='cuda:0'), 'train_loss_bbox': tensor(0.0246, device='cuda:0'), 'train_loss_giou': tensor(0.1206, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.7050, device='cuda:0'), 'validation_loss_ce': tensor(0.4017, device='cuda:0'), 'validation_loss_bbox': tensor(0.0211, device='cuda:0'), 'validation_loss_giou': tensor(0.0988, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.7050, device='cuda:0'), 'train_loss_ce': tensor(0.4017, device='cuda:0'), 'train_loss_bbox': tensor(0.0211, device='cuda:0'), 'train_loss_giou': tensor(0.0988, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6875, device='cuda:0'), 'validation_loss_ce': tensor(0.4006, device='cuda:0'), 'validation_loss_bbox': tensor(0.0190, device='cuda:0'), 'validation_loss_giou': tensor(0.0961, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6875, device='cuda:0'), 'train_loss_ce': tensor(0.4006, device='cuda:0'), 'train_loss_bbox': tensor(0.0190, device='cuda:0'), 'train_loss_giou': tensor(0.0961, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5480, device='cuda:0'), 'validation_loss_ce': tensor(0.3992, device='cuda:0'), 'validation_loss_bbox': tensor(0.0104, device='cuda:0'), 'validation_loss_giou': tensor(0.0484, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5480, device='cuda:0'), 'train_loss_ce': tensor(0.3992, device='cuda:0'), 'train_loss_bbox': tensor(0.0104, device='cuda:0'), 'train_loss_giou': tensor(0.0484, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5450, device='cuda:0'), 'validation_loss_ce': tensor(0.3983, device='cuda:0'), 'validation_loss_bbox': tensor(0.0099, device='cuda:0'), 'validation_loss_giou': tensor(0.0486, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5450, device='cuda:0'), 'train_loss_ce': tensor(0.3983, device='cuda:0'), 'train_loss_bbox': tensor(0.0099, device='cuda:0'), 'train_loss_giou': tensor(0.0486, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6349, device='cuda:0'), 'validation_loss_ce': tensor(0.3971, device='cuda:0'), 'validation_loss_bbox': tensor(0.0148, device='cuda:0'), 'validation_loss_giou': tensor(0.0819, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6349, device='cuda:0'), 'train_loss_ce': tensor(0.3971, device='cuda:0'), 'train_loss_bbox': tensor(0.0148, device='cuda:0'), 'train_loss_giou': tensor(0.0819, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6265, device='cuda:0'), 'validation_loss_ce': tensor(0.3958, device='cuda:0'), 'validation_loss_bbox': tensor(0.0169, device='cuda:0'), 'validation_loss_giou': tensor(0.0730, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6265, device='cuda:0'), 'train_loss_ce': tensor(0.3958, device='cuda:0'), 'train_loss_bbox': tensor(0.0169, device='cuda:0'), 'train_loss_giou': tensor(0.0730, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6520, device='cuda:0'), 'validation_loss_ce': tensor(0.3946, device='cuda:0'), 'validation_loss_bbox': tensor(0.0162, device='cuda:0'), 'validation_loss_giou': tensor(0.0881, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6520, device='cuda:0'), 'train_loss_ce': tensor(0.3946, device='cuda:0'), 'train_loss_bbox': tensor(0.0162, device='cuda:0'), 'train_loss_giou': tensor(0.0881, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6477, device='cuda:0'), 'validation_loss_ce': tensor(0.3935, device='cuda:0'), 'validation_loss_bbox': tensor(0.0166, device='cuda:0'), 'validation_loss_giou': tensor(0.0856, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6477, device='cuda:0'), 'train_loss_ce': tensor(0.3935, device='cuda:0'), 'train_loss_bbox': tensor(0.0166, device='cuda:0'), 'train_loss_giou': tensor(0.0856, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5710, device='cuda:0'), 'validation_loss_ce': tensor(0.3923, device='cuda:0'), 'validation_loss_bbox': tensor(0.0134, device='cuda:0'), 'validation_loss_giou': tensor(0.0559, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5710, device='cuda:0'), 'train_loss_ce': tensor(0.3923, device='cuda:0'), 'train_loss_bbox': tensor(0.0134, device='cuda:0'), 'train_loss_giou': tensor(0.0559, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6067, device='cuda:0'), 'validation_loss_ce': tensor(0.3910, device='cuda:0'), 'validation_loss_bbox': tensor(0.0105, device='cuda:0'), 'validation_loss_giou': tensor(0.0818, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6067, device='cuda:0'), 'train_loss_ce': tensor(0.3910, device='cuda:0'), 'train_loss_bbox': tensor(0.0105, device='cuda:0'), 'train_loss_giou': tensor(0.0818, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5941, device='cuda:0'), 'validation_loss_ce': tensor(0.3898, device='cuda:0'), 'validation_loss_bbox': tensor(0.0097, device='cuda:0'), 'validation_loss_giou': tensor(0.0778, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5941, device='cuda:0'), 'train_loss_ce': tensor(0.3898, device='cuda:0'), 'train_loss_bbox': tensor(0.0097, device='cuda:0'), 'train_loss_giou': tensor(0.0778, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6495, device='cuda:0'), 'validation_loss_ce': tensor(0.3890, device='cuda:0'), 'validation_loss_bbox': tensor(0.0139, device='cuda:0'), 'validation_loss_giou': tensor(0.0956, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6495, device='cuda:0'), 'train_loss_ce': tensor(0.3890, device='cuda:0'), 'train_loss_bbox': tensor(0.0139, device='cuda:0'), 'train_loss_giou': tensor(0.0956, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6510, device='cuda:0'), 'validation_loss_ce': tensor(0.3882, device='cuda:0'), 'validation_loss_bbox': tensor(0.0147, device='cuda:0'), 'validation_loss_giou': tensor(0.0947, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6510, device='cuda:0'), 'train_loss_ce': tensor(0.3882, device='cuda:0'), 'train_loss_bbox': tensor(0.0147, device='cuda:0'), 'train_loss_giou': tensor(0.0947, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6475, device='cuda:0'), 'validation_loss_ce': tensor(0.3872, device='cuda:0'), 'validation_loss_bbox': tensor(0.0162, device='cuda:0'), 'validation_loss_giou': tensor(0.0895, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6475, device='cuda:0'), 'train_loss_ce': tensor(0.3872, device='cuda:0'), 'train_loss_bbox': tensor(0.0162, device='cuda:0'), 'train_loss_giou': tensor(0.0895, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5954, device='cuda:0'), 'validation_loss_ce': tensor(0.3859, device='cuda:0'), 'validation_loss_bbox': tensor(0.0125, device='cuda:0'), 'validation_loss_giou': tensor(0.0735, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5954, device='cuda:0'), 'train_loss_ce': tensor(0.3859, device='cuda:0'), 'train_loss_bbox': tensor(0.0125, device='cuda:0'), 'train_loss_giou': tensor(0.0735, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5807, device='cuda:0'), 'validation_loss_ce': tensor(0.3845, device='cuda:0'), 'validation_loss_bbox': tensor(0.0146, device='cuda:0'), 'validation_loss_giou': tensor(0.0615, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5807, device='cuda:0'), 'train_loss_ce': tensor(0.3845, device='cuda:0'), 'train_loss_bbox': tensor(0.0146, device='cuda:0'), 'train_loss_giou': tensor(0.0615, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5899, device='cuda:0'), 'validation_loss_ce': tensor(0.3832, device='cuda:0'), 'validation_loss_bbox': tensor(0.0145, device='cuda:0'), 'validation_loss_giou': tensor(0.0671, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5899, device='cuda:0'), 'train_loss_ce': tensor(0.3832, device='cuda:0'), 'train_loss_bbox': tensor(0.0145, device='cuda:0'), 'train_loss_giou': tensor(0.0671, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6029, device='cuda:0'), 'validation_loss_ce': tensor(0.3822, device='cuda:0'), 'validation_loss_bbox': tensor(0.0146, device='cuda:0'), 'validation_loss_giou': tensor(0.0738, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6029, device='cuda:0'), 'train_loss_ce': tensor(0.3822, device='cuda:0'), 'train_loss_bbox': tensor(0.0146, device='cuda:0'), 'train_loss_giou': tensor(0.0738, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5559, device='cuda:0'), 'validation_loss_ce': tensor(0.3813, device='cuda:0'), 'validation_loss_bbox': tensor(0.0113, device='cuda:0'), 'validation_loss_giou': tensor(0.0592, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5559, device='cuda:0'), 'train_loss_ce': tensor(0.3813, device='cuda:0'), 'train_loss_bbox': tensor(0.0113, device='cuda:0'), 'train_loss_giou': tensor(0.0592, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5969, device='cuda:0'), 'validation_loss_ce': tensor(0.3805, device='cuda:0'), 'validation_loss_bbox': tensor(0.0096, device='cuda:0'), 'validation_loss_giou': tensor(0.0842, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5969, device='cuda:0'), 'train_loss_ce': tensor(0.3805, device='cuda:0'), 'train_loss_bbox': tensor(0.0096, device='cuda:0'), 'train_loss_giou': tensor(0.0842, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6054, device='cuda:0'), 'validation_loss_ce': tensor(0.3789, device='cuda:0'), 'validation_loss_bbox': tensor(0.0117, device='cuda:0'), 'validation_loss_giou': tensor(0.0841, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6054, device='cuda:0'), 'train_loss_ce': tensor(0.3789, device='cuda:0'), 'train_loss_bbox': tensor(0.0117, device='cuda:0'), 'train_loss_giou': tensor(0.0841, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5378, device='cuda:0'), 'validation_loss_ce': tensor(0.3779, device='cuda:0'), 'validation_loss_bbox': tensor(0.0100, device='cuda:0'), 'validation_loss_giou': tensor(0.0549, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5378, device='cuda:0'), 'train_loss_ce': tensor(0.3779, device='cuda:0'), 'train_loss_bbox': tensor(0.0100, device='cuda:0'), 'train_loss_giou': tensor(0.0549, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4806, device='cuda:0'), 'validation_loss_ce': tensor(0.3766, device='cuda:0'), 'validation_loss_bbox': tensor(0.0071, device='cuda:0'), 'validation_loss_giou': tensor(0.0342, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4806, device='cuda:0'), 'train_loss_ce': tensor(0.3766, device='cuda:0'), 'train_loss_bbox': tensor(0.0071, device='cuda:0'), 'train_loss_giou': tensor(0.0342, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5535, device='cuda:0'), 'validation_loss_ce': tensor(0.3754, device='cuda:0'), 'validation_loss_bbox': tensor(0.0126, device='cuda:0'), 'validation_loss_giou': tensor(0.0576, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5535, device='cuda:0'), 'train_loss_ce': tensor(0.3754, device='cuda:0'), 'train_loss_bbox': tensor(0.0126, device='cuda:0'), 'train_loss_giou': tensor(0.0576, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5231, device='cuda:0'), 'validation_loss_ce': tensor(0.3745, device='cuda:0'), 'validation_loss_bbox': tensor(0.0117, device='cuda:0'), 'validation_loss_giou': tensor(0.0450, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5231, device='cuda:0'), 'train_loss_ce': tensor(0.3745, device='cuda:0'), 'train_loss_bbox': tensor(0.0117, device='cuda:0'), 'train_loss_giou': tensor(0.0450, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5471, device='cuda:0'), 'validation_loss_ce': tensor(0.3739, device='cuda:0'), 'validation_loss_bbox': tensor(0.0101, device='cuda:0'), 'validation_loss_giou': tensor(0.0614, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5471, device='cuda:0'), 'train_loss_ce': tensor(0.3739, device='cuda:0'), 'train_loss_bbox': tensor(0.0101, device='cuda:0'), 'train_loss_giou': tensor(0.0614, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5491, device='cuda:0'), 'validation_loss_ce': tensor(0.3733, device='cuda:0'), 'validation_loss_bbox': tensor(0.0104, device='cuda:0'), 'validation_loss_giou': tensor(0.0619, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5491, device='cuda:0'), 'train_loss_ce': tensor(0.3733, device='cuda:0'), 'train_loss_bbox': tensor(0.0104, device='cuda:0'), 'train_loss_giou': tensor(0.0619, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5884, device='cuda:0'), 'validation_loss_ce': tensor(0.3727, device='cuda:0'), 'validation_loss_bbox': tensor(0.0111, device='cuda:0'), 'validation_loss_giou': tensor(0.0802, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5884, device='cuda:0'), 'train_loss_ce': tensor(0.3727, device='cuda:0'), 'train_loss_bbox': tensor(0.0111, device='cuda:0'), 'train_loss_giou': tensor(0.0802, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5379, device='cuda:0'), 'validation_loss_ce': tensor(0.3719, device='cuda:0'), 'validation_loss_bbox': tensor(0.0099, device='cuda:0'), 'validation_loss_giou': tensor(0.0581, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5379, device='cuda:0'), 'train_loss_ce': tensor(0.3719, device='cuda:0'), 'train_loss_bbox': tensor(0.0099, device='cuda:0'), 'train_loss_giou': tensor(0.0581, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5009, device='cuda:0'), 'validation_loss_ce': tensor(0.3710, device='cuda:0'), 'validation_loss_bbox': tensor(0.0080, device='cuda:0'), 'validation_loss_giou': tensor(0.0450, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5009, device='cuda:0'), 'train_loss_ce': tensor(0.3710, device='cuda:0'), 'train_loss_bbox': tensor(0.0080, device='cuda:0'), 'train_loss_giou': tensor(0.0450, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5166, device='cuda:0'), 'validation_loss_ce': tensor(0.3701, device='cuda:0'), 'validation_loss_bbox': tensor(0.0078, device='cuda:0'), 'validation_loss_giou': tensor(0.0538, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5166, device='cuda:0'), 'train_loss_ce': tensor(0.3701, device='cuda:0'), 'train_loss_bbox': tensor(0.0078, device='cuda:0'), 'train_loss_giou': tensor(0.0538, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5222, device='cuda:0'), 'validation_loss_ce': tensor(0.3693, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0540, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5222, device='cuda:0'), 'train_loss_ce': tensor(0.3693, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0540, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5259, device='cuda:0'), 'validation_loss_ce': tensor(0.3685, device='cuda:0'), 'validation_loss_bbox': tensor(0.0091, device='cuda:0'), 'validation_loss_giou': tensor(0.0559, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5259, device='cuda:0'), 'train_loss_ce': tensor(0.3685, device='cuda:0'), 'train_loss_bbox': tensor(0.0091, device='cuda:0'), 'train_loss_giou': tensor(0.0559, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5014, device='cuda:0'), 'validation_loss_ce': tensor(0.3678, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0442, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5014, device='cuda:0'), 'train_loss_ce': tensor(0.3678, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0442, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4879, device='cuda:0'), 'validation_loss_ce': tensor(0.3671, device='cuda:0'), 'validation_loss_bbox': tensor(0.0079, device='cuda:0'), 'validation_loss_giou': tensor(0.0405, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4879, device='cuda:0'), 'train_loss_ce': tensor(0.3671, device='cuda:0'), 'train_loss_bbox': tensor(0.0079, device='cuda:0'), 'train_loss_giou': tensor(0.0405, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5354, device='cuda:0'), 'validation_loss_ce': tensor(0.3662, device='cuda:0'), 'validation_loss_bbox': tensor(0.0111, device='cuda:0'), 'validation_loss_giou': tensor(0.0567, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5354, device='cuda:0'), 'train_loss_ce': tensor(0.3662, device='cuda:0'), 'train_loss_bbox': tensor(0.0111, device='cuda:0'), 'train_loss_giou': tensor(0.0567, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5425, device='cuda:0'), 'validation_loss_ce': tensor(0.3654, device='cuda:0'), 'validation_loss_bbox': tensor(0.0094, device='cuda:0'), 'validation_loss_giou': tensor(0.0650, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5425, device='cuda:0'), 'train_loss_ce': tensor(0.3654, device='cuda:0'), 'train_loss_bbox': tensor(0.0094, device='cuda:0'), 'train_loss_giou': tensor(0.0650, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6095, device='cuda:0'), 'validation_loss_ce': tensor(0.3646, device='cuda:0'), 'validation_loss_bbox': tensor(0.0144, device='cuda:0'), 'validation_loss_giou': tensor(0.0865, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6095, device='cuda:0'), 'train_loss_ce': tensor(0.3646, device='cuda:0'), 'train_loss_bbox': tensor(0.0144, device='cuda:0'), 'train_loss_giou': tensor(0.0865, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5793, device='cuda:0'), 'validation_loss_ce': tensor(0.3637, device='cuda:0'), 'validation_loss_bbox': tensor(0.0123, device='cuda:0'), 'validation_loss_giou': tensor(0.0771, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5793, device='cuda:0'), 'train_loss_ce': tensor(0.3637, device='cuda:0'), 'train_loss_bbox': tensor(0.0123, device='cuda:0'), 'train_loss_giou': tensor(0.0771, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5058, device='cuda:0'), 'validation_loss_ce': tensor(0.3627, device='cuda:0'), 'validation_loss_bbox': tensor(0.0094, device='cuda:0'), 'validation_loss_giou': tensor(0.0481, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5058, device='cuda:0'), 'train_loss_ce': tensor(0.3627, device='cuda:0'), 'train_loss_bbox': tensor(0.0094, device='cuda:0'), 'train_loss_giou': tensor(0.0481, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5653, device='cuda:0'), 'validation_loss_ce': tensor(0.3619, device='cuda:0'), 'validation_loss_bbox': tensor(0.0106, device='cuda:0'), 'validation_loss_giou': tensor(0.0753, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5653, device='cuda:0'), 'train_loss_ce': tensor(0.3619, device='cuda:0'), 'train_loss_bbox': tensor(0.0106, device='cuda:0'), 'train_loss_giou': tensor(0.0753, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5217, device='cuda:0'), 'validation_loss_ce': tensor(0.3612, device='cuda:0'), 'validation_loss_bbox': tensor(0.0124, device='cuda:0'), 'validation_loss_giou': tensor(0.0493, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5217, device='cuda:0'), 'train_loss_ce': tensor(0.3612, device='cuda:0'), 'train_loss_bbox': tensor(0.0124, device='cuda:0'), 'train_loss_giou': tensor(0.0493, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5408, device='cuda:0'), 'validation_loss_ce': tensor(0.3605, device='cuda:0'), 'validation_loss_bbox': tensor(0.0118, device='cuda:0'), 'validation_loss_giou': tensor(0.0607, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5408, device='cuda:0'), 'train_loss_ce': tensor(0.3605, device='cuda:0'), 'train_loss_bbox': tensor(0.0118, device='cuda:0'), 'train_loss_giou': tensor(0.0607, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5807, device='cuda:0'), 'validation_loss_ce': tensor(0.3597, device='cuda:0'), 'validation_loss_bbox': tensor(0.0159, device='cuda:0'), 'validation_loss_giou': tensor(0.0707, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5807, device='cuda:0'), 'train_loss_ce': tensor(0.3597, device='cuda:0'), 'train_loss_bbox': tensor(0.0159, device='cuda:0'), 'train_loss_giou': tensor(0.0707, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5712, device='cuda:0'), 'validation_loss_ce': tensor(0.3591, device='cuda:0'), 'validation_loss_bbox': tensor(0.0134, device='cuda:0'), 'validation_loss_giou': tensor(0.0727, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5712, device='cuda:0'), 'train_loss_ce': tensor(0.3591, device='cuda:0'), 'train_loss_bbox': tensor(0.0134, device='cuda:0'), 'train_loss_giou': tensor(0.0727, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4965, device='cuda:0'), 'validation_loss_ce': tensor(0.3585, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0464, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4965, device='cuda:0'), 'train_loss_ce': tensor(0.3585, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0464, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5830, device='cuda:0'), 'validation_loss_ce': tensor(0.3579, device='cuda:0'), 'validation_loss_bbox': tensor(0.0109, device='cuda:0'), 'validation_loss_giou': tensor(0.0853, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5830, device='cuda:0'), 'train_loss_ce': tensor(0.3579, device='cuda:0'), 'train_loss_bbox': tensor(0.0109, device='cuda:0'), 'train_loss_giou': tensor(0.0853, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5602, device='cuda:0'), 'validation_loss_ce': tensor(0.3571, device='cuda:0'), 'validation_loss_bbox': tensor(0.0092, device='cuda:0'), 'validation_loss_giou': tensor(0.0786, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5602, device='cuda:0'), 'train_loss_ce': tensor(0.3571, device='cuda:0'), 'train_loss_bbox': tensor(0.0092, device='cuda:0'), 'train_loss_giou': tensor(0.0786, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5277, device='cuda:0'), 'validation_loss_ce': tensor(0.3565, device='cuda:0'), 'validation_loss_bbox': tensor(0.0095, device='cuda:0'), 'validation_loss_giou': tensor(0.0619, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5277, device='cuda:0'), 'train_loss_ce': tensor(0.3565, device='cuda:0'), 'train_loss_bbox': tensor(0.0095, device='cuda:0'), 'train_loss_giou': tensor(0.0619, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5530, device='cuda:0'), 'validation_loss_ce': tensor(0.3557, device='cuda:0'), 'validation_loss_bbox': tensor(0.0105, device='cuda:0'), 'validation_loss_giou': tensor(0.0725, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5530, device='cuda:0'), 'train_loss_ce': tensor(0.3557, device='cuda:0'), 'train_loss_bbox': tensor(0.0105, device='cuda:0'), 'train_loss_giou': tensor(0.0725, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5450, device='cuda:0'), 'validation_loss_ce': tensor(0.3545, device='cuda:0'), 'validation_loss_bbox': tensor(0.0104, device='cuda:0'), 'validation_loss_giou': tensor(0.0693, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5450, device='cuda:0'), 'train_loss_ce': tensor(0.3545, device='cuda:0'), 'train_loss_bbox': tensor(0.0104, device='cuda:0'), 'train_loss_giou': tensor(0.0693, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4931, device='cuda:0'), 'validation_loss_ce': tensor(0.3538, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0472, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4931, device='cuda:0'), 'train_loss_ce': tensor(0.3538, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0472, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5782, device='cuda:0'), 'validation_loss_ce': tensor(0.3530, device='cuda:0'), 'validation_loss_bbox': tensor(0.0124, device='cuda:0'), 'validation_loss_giou': tensor(0.0815, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5782, device='cuda:0'), 'train_loss_ce': tensor(0.3530, device='cuda:0'), 'train_loss_bbox': tensor(0.0124, device='cuda:0'), 'train_loss_giou': tensor(0.0815, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5109, device='cuda:0'), 'validation_loss_ce': tensor(0.3522, device='cuda:0'), 'validation_loss_bbox': tensor(0.0101, device='cuda:0'), 'validation_loss_giou': tensor(0.0542, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5109, device='cuda:0'), 'train_loss_ce': tensor(0.3522, device='cuda:0'), 'train_loss_bbox': tensor(0.0101, device='cuda:0'), 'train_loss_giou': tensor(0.0542, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5714, device='cuda:0'), 'validation_loss_ce': tensor(0.3515, device='cuda:0'), 'validation_loss_bbox': tensor(0.0120, device='cuda:0'), 'validation_loss_giou': tensor(0.0800, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5714, device='cuda:0'), 'train_loss_ce': tensor(0.3515, device='cuda:0'), 'train_loss_bbox': tensor(0.0120, device='cuda:0'), 'train_loss_giou': tensor(0.0800, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6086, device='cuda:0'), 'validation_loss_ce': tensor(0.3510, device='cuda:0'), 'validation_loss_bbox': tensor(0.0139, device='cuda:0'), 'validation_loss_giou': tensor(0.0942, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6086, device='cuda:0'), 'train_loss_ce': tensor(0.3510, device='cuda:0'), 'train_loss_bbox': tensor(0.0139, device='cuda:0'), 'train_loss_giou': tensor(0.0942, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4965, device='cuda:0'), 'validation_loss_ce': tensor(0.3505, device='cuda:0'), 'validation_loss_bbox': tensor(0.0103, device='cuda:0'), 'validation_loss_giou': tensor(0.0472, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4965, device='cuda:0'), 'train_loss_ce': tensor(0.3505, device='cuda:0'), 'train_loss_bbox': tensor(0.0103, device='cuda:0'), 'train_loss_giou': tensor(0.0472, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6358, device='cuda:0'), 'validation_loss_ce': tensor(0.3498, device='cuda:0'), 'validation_loss_bbox': tensor(0.0126, device='cuda:0'), 'validation_loss_giou': tensor(0.1116, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6358, device='cuda:0'), 'train_loss_ce': tensor(0.3498, device='cuda:0'), 'train_loss_bbox': tensor(0.0126, device='cuda:0'), 'train_loss_giou': tensor(0.1116, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6799, device='cuda:0'), 'validation_loss_ce': tensor(0.3488, device='cuda:0'), 'validation_loss_bbox': tensor(0.0155, device='cuda:0'), 'validation_loss_giou': tensor(0.1268, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6799, device='cuda:0'), 'train_loss_ce': tensor(0.3488, device='cuda:0'), 'train_loss_bbox': tensor(0.0155, device='cuda:0'), 'train_loss_giou': tensor(0.1268, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6138, device='cuda:0'), 'validation_loss_ce': tensor(0.3479, device='cuda:0'), 'validation_loss_bbox': tensor(0.0131, device='cuda:0'), 'validation_loss_giou': tensor(0.1003, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6138, device='cuda:0'), 'train_loss_ce': tensor(0.3479, device='cuda:0'), 'train_loss_bbox': tensor(0.0131, device='cuda:0'), 'train_loss_giou': tensor(0.1003, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5149, device='cuda:0'), 'validation_loss_ce': tensor(0.3471, device='cuda:0'), 'validation_loss_bbox': tensor(0.0095, device='cuda:0'), 'validation_loss_giou': tensor(0.0600, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5149, device='cuda:0'), 'train_loss_ce': tensor(0.3471, device='cuda:0'), 'train_loss_bbox': tensor(0.0095, device='cuda:0'), 'train_loss_giou': tensor(0.0600, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5965, device='cuda:0'), 'validation_loss_ce': tensor(0.3466, device='cuda:0'), 'validation_loss_bbox': tensor(0.0128, device='cuda:0'), 'validation_loss_giou': tensor(0.0929, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5965, device='cuda:0'), 'train_loss_ce': tensor(0.3466, device='cuda:0'), 'train_loss_bbox': tensor(0.0128, device='cuda:0'), 'train_loss_giou': tensor(0.0929, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6316, device='cuda:0'), 'validation_loss_ce': tensor(0.3459, device='cuda:0'), 'validation_loss_bbox': tensor(0.0116, device='cuda:0'), 'validation_loss_giou': tensor(0.1139, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6316, device='cuda:0'), 'train_loss_ce': tensor(0.3459, device='cuda:0'), 'train_loss_bbox': tensor(0.0116, device='cuda:0'), 'train_loss_giou': tensor(0.1139, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6073, device='cuda:0'), 'validation_loss_ce': tensor(0.3453, device='cuda:0'), 'validation_loss_bbox': tensor(0.0116, device='cuda:0'), 'validation_loss_giou': tensor(0.1020, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6073, device='cuda:0'), 'train_loss_ce': tensor(0.3453, device='cuda:0'), 'train_loss_bbox': tensor(0.0116, device='cuda:0'), 'train_loss_giou': tensor(0.1020, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5714, device='cuda:0'), 'validation_loss_ce': tensor(0.3445, device='cuda:0'), 'validation_loss_bbox': tensor(0.0118, device='cuda:0'), 'validation_loss_giou': tensor(0.0839, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5714, device='cuda:0'), 'train_loss_ce': tensor(0.3445, device='cuda:0'), 'train_loss_bbox': tensor(0.0118, device='cuda:0'), 'train_loss_giou': tensor(0.0839, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5379, device='cuda:0'), 'validation_loss_ce': tensor(0.3437, device='cuda:0'), 'validation_loss_bbox': tensor(0.0120, device='cuda:0'), 'validation_loss_giou': tensor(0.0672, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5379, device='cuda:0'), 'train_loss_ce': tensor(0.3437, device='cuda:0'), 'train_loss_bbox': tensor(0.0120, device='cuda:0'), 'train_loss_giou': tensor(0.0672, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5909, device='cuda:0'), 'validation_loss_ce': tensor(0.3429, device='cuda:0'), 'validation_loss_bbox': tensor(0.0123, device='cuda:0'), 'validation_loss_giou': tensor(0.0932, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5909, device='cuda:0'), 'train_loss_ce': tensor(0.3429, device='cuda:0'), 'train_loss_bbox': tensor(0.0123, device='cuda:0'), 'train_loss_giou': tensor(0.0932, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4832, device='cuda:0'), 'validation_loss_ce': tensor(0.3424, device='cuda:0'), 'validation_loss_bbox': tensor(0.0070, device='cuda:0'), 'validation_loss_giou': tensor(0.0529, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4832, device='cuda:0'), 'train_loss_ce': tensor(0.3424, device='cuda:0'), 'train_loss_bbox': tensor(0.0070, device='cuda:0'), 'train_loss_giou': tensor(0.0529, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5717, device='cuda:0'), 'validation_loss_ce': tensor(0.3419, device='cuda:0'), 'validation_loss_bbox': tensor(0.0134, device='cuda:0'), 'validation_loss_giou': tensor(0.0814, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5717, device='cuda:0'), 'train_loss_ce': tensor(0.3419, device='cuda:0'), 'train_loss_bbox': tensor(0.0134, device='cuda:0'), 'train_loss_giou': tensor(0.0814, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.6517, device='cuda:0'), 'validation_loss_ce': tensor(0.3413, device='cuda:0'), 'validation_loss_bbox': tensor(0.0184, device='cuda:0'), 'validation_loss_giou': tensor(0.1093, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.6517, device='cuda:0'), 'train_loss_ce': tensor(0.3413, device='cuda:0'), 'train_loss_bbox': tensor(0.0184, device='cuda:0'), 'train_loss_giou': tensor(0.1093, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5885, device='cuda:0'), 'validation_loss_ce': tensor(0.3405, device='cuda:0'), 'validation_loss_bbox': tensor(0.0154, device='cuda:0'), 'validation_loss_giou': tensor(0.0855, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5885, device='cuda:0'), 'train_loss_ce': tensor(0.3405, device='cuda:0'), 'train_loss_bbox': tensor(0.0154, device='cuda:0'), 'train_loss_giou': tensor(0.0855, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4941, device='cuda:0'), 'validation_loss_ce': tensor(0.3397, device='cuda:0'), 'validation_loss_bbox': tensor(0.0094, device='cuda:0'), 'validation_loss_giou': tensor(0.0536, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4941, device='cuda:0'), 'train_loss_ce': tensor(0.3397, device='cuda:0'), 'train_loss_bbox': tensor(0.0094, device='cuda:0'), 'train_loss_giou': tensor(0.0536, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4940, device='cuda:0'), 'validation_loss_ce': tensor(0.3389, device='cuda:0'), 'validation_loss_bbox': tensor(0.0078, device='cuda:0'), 'validation_loss_giou': tensor(0.0581, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4940, device='cuda:0'), 'train_loss_ce': tensor(0.3389, device='cuda:0'), 'train_loss_bbox': tensor(0.0078, device='cuda:0'), 'train_loss_giou': tensor(0.0581, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4508, device='cuda:0'), 'validation_loss_ce': tensor(0.3382, device='cuda:0'), 'validation_loss_bbox': tensor(0.0078, device='cuda:0'), 'validation_loss_giou': tensor(0.0367, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4508, device='cuda:0'), 'train_loss_ce': tensor(0.3382, device='cuda:0'), 'train_loss_bbox': tensor(0.0078, device='cuda:0'), 'train_loss_giou': tensor(0.0367, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4941, device='cuda:0'), 'validation_loss_ce': tensor(0.3375, device='cuda:0'), 'validation_loss_bbox': tensor(0.0078, device='cuda:0'), 'validation_loss_giou': tensor(0.0588, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4941, device='cuda:0'), 'train_loss_ce': tensor(0.3375, device='cuda:0'), 'train_loss_bbox': tensor(0.0078, device='cuda:0'), 'train_loss_giou': tensor(0.0588, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4695, device='cuda:0'), 'validation_loss_ce': tensor(0.3366, device='cuda:0'), 'validation_loss_bbox': tensor(0.0077, device='cuda:0'), 'validation_loss_giou': tensor(0.0471, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4695, device='cuda:0'), 'train_loss_ce': tensor(0.3366, device='cuda:0'), 'train_loss_bbox': tensor(0.0077, device='cuda:0'), 'train_loss_giou': tensor(0.0471, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5122, device='cuda:0'), 'validation_loss_ce': tensor(0.3355, device='cuda:0'), 'validation_loss_bbox': tensor(0.0084, device='cuda:0'), 'validation_loss_giou': tensor(0.0672, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5122, device='cuda:0'), 'train_loss_ce': tensor(0.3355, device='cuda:0'), 'train_loss_bbox': tensor(0.0084, device='cuda:0'), 'train_loss_giou': tensor(0.0672, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5361, device='cuda:0'), 'validation_loss_ce': tensor(0.3347, device='cuda:0'), 'validation_loss_bbox': tensor(0.0096, device='cuda:0'), 'validation_loss_giou': tensor(0.0768, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5361, device='cuda:0'), 'train_loss_ce': tensor(0.3347, device='cuda:0'), 'train_loss_bbox': tensor(0.0096, device='cuda:0'), 'train_loss_giou': tensor(0.0768, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4663, device='cuda:0'), 'validation_loss_ce': tensor(0.3340, device='cuda:0'), 'validation_loss_bbox': tensor(0.0067, device='cuda:0'), 'validation_loss_giou': tensor(0.0494, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4663, device='cuda:0'), 'train_loss_ce': tensor(0.3340, device='cuda:0'), 'train_loss_bbox': tensor(0.0067, device='cuda:0'), 'train_loss_giou': tensor(0.0494, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4465, device='cuda:0'), 'validation_loss_ce': tensor(0.3335, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0391, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4465, device='cuda:0'), 'train_loss_ce': tensor(0.3335, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0391, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5186, device='cuda:0'), 'validation_loss_ce': tensor(0.3331, device='cuda:0'), 'validation_loss_bbox': tensor(0.0091, device='cuda:0'), 'validation_loss_giou': tensor(0.0700, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5186, device='cuda:0'), 'train_loss_ce': tensor(0.3331, device='cuda:0'), 'train_loss_bbox': tensor(0.0091, device='cuda:0'), 'train_loss_giou': tensor(0.0700, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5086, device='cuda:0'), 'validation_loss_ce': tensor(0.3325, device='cuda:0'), 'validation_loss_bbox': tensor(0.0091, device='cuda:0'), 'validation_loss_giou': tensor(0.0655, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5086, device='cuda:0'), 'train_loss_ce': tensor(0.3325, device='cuda:0'), 'train_loss_bbox': tensor(0.0091, device='cuda:0'), 'train_loss_giou': tensor(0.0655, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4515, device='cuda:0'), 'validation_loss_ce': tensor(0.3317, device='cuda:0'), 'validation_loss_bbox': tensor(0.0076, device='cuda:0'), 'validation_loss_giou': tensor(0.0410, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4515, device='cuda:0'), 'train_loss_ce': tensor(0.3317, device='cuda:0'), 'train_loss_bbox': tensor(0.0076, device='cuda:0'), 'train_loss_giou': tensor(0.0410, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4460, device='cuda:0'), 'validation_loss_ce': tensor(0.3309, device='cuda:0'), 'validation_loss_bbox': tensor(0.0064, device='cuda:0'), 'validation_loss_giou': tensor(0.0415, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4460, device='cuda:0'), 'train_loss_ce': tensor(0.3309, device='cuda:0'), 'train_loss_bbox': tensor(0.0064, device='cuda:0'), 'train_loss_giou': tensor(0.0415, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4363, device='cuda:0'), 'validation_loss_ce': tensor(0.3302, device='cuda:0'), 'validation_loss_bbox': tensor(0.0074, device='cuda:0'), 'validation_loss_giou': tensor(0.0346, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4363, device='cuda:0'), 'train_loss_ce': tensor(0.3302, device='cuda:0'), 'train_loss_bbox': tensor(0.0074, device='cuda:0'), 'train_loss_giou': tensor(0.0346, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4712, device='cuda:0'), 'validation_loss_ce': tensor(0.3314, device='cuda:0'), 'validation_loss_bbox': tensor(0.0077, device='cuda:0'), 'validation_loss_giou': tensor(0.0506, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4712, device='cuda:0'), 'train_loss_ce': tensor(0.3314, device='cuda:0'), 'train_loss_bbox': tensor(0.0077, device='cuda:0'), 'train_loss_giou': tensor(0.0506, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4224, device='cuda:0'), 'validation_loss_ce': tensor(0.3305, device='cuda:0'), 'validation_loss_bbox': tensor(0.0070, device='cuda:0'), 'validation_loss_giou': tensor(0.0283, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4224, device='cuda:0'), 'train_loss_ce': tensor(0.3305, device='cuda:0'), 'train_loss_bbox': tensor(0.0070, device='cuda:0'), 'train_loss_giou': tensor(0.0283, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4075, device='cuda:0'), 'validation_loss_ce': tensor(0.3296, device='cuda:0'), 'validation_loss_bbox': tensor(0.0057, device='cuda:0'), 'validation_loss_giou': tensor(0.0246, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4075, device='cuda:0'), 'train_loss_ce': tensor(0.3296, device='cuda:0'), 'train_loss_bbox': tensor(0.0057, device='cuda:0'), 'train_loss_giou': tensor(0.0246, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4404, device='cuda:0'), 'validation_loss_ce': tensor(0.3289, device='cuda:0'), 'validation_loss_bbox': tensor(0.0062, device='cuda:0'), 'validation_loss_giou': tensor(0.0402, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4404, device='cuda:0'), 'train_loss_ce': tensor(0.3289, device='cuda:0'), 'train_loss_bbox': tensor(0.0062, device='cuda:0'), 'train_loss_giou': tensor(0.0402, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4306, device='cuda:0'), 'validation_loss_ce': tensor(0.3281, device='cuda:0'), 'validation_loss_bbox': tensor(0.0073, device='cuda:0'), 'validation_loss_giou': tensor(0.0330, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4306, device='cuda:0'), 'train_loss_ce': tensor(0.3281, device='cuda:0'), 'train_loss_bbox': tensor(0.0073, device='cuda:0'), 'train_loss_giou': tensor(0.0330, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5279, device='cuda:0'), 'validation_loss_ce': tensor(0.3274, device='cuda:0'), 'validation_loss_bbox': tensor(0.0103, device='cuda:0'), 'validation_loss_giou': tensor(0.0745, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5279, device='cuda:0'), 'train_loss_ce': tensor(0.3274, device='cuda:0'), 'train_loss_bbox': tensor(0.0103, device='cuda:0'), 'train_loss_giou': tensor(0.0745, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4982, device='cuda:0'), 'validation_loss_ce': tensor(0.3267, device='cuda:0'), 'validation_loss_bbox': tensor(0.0086, device='cuda:0'), 'validation_loss_giou': tensor(0.0642, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4982, device='cuda:0'), 'train_loss_ce': tensor(0.3267, device='cuda:0'), 'train_loss_bbox': tensor(0.0086, device='cuda:0'), 'train_loss_giou': tensor(0.0642, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4468, device='cuda:0'), 'validation_loss_ce': tensor(0.3261, device='cuda:0'), 'validation_loss_bbox': tensor(0.0082, device='cuda:0'), 'validation_loss_giou': tensor(0.0398, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4468, device='cuda:0'), 'train_loss_ce': tensor(0.3261, device='cuda:0'), 'train_loss_bbox': tensor(0.0082, device='cuda:0'), 'train_loss_giou': tensor(0.0398, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5239, device='cuda:0'), 'validation_loss_ce': tensor(0.3254, device='cuda:0'), 'validation_loss_bbox': tensor(0.0119, device='cuda:0'), 'validation_loss_giou': tensor(0.0696, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5239, device='cuda:0'), 'train_loss_ce': tensor(0.3254, device='cuda:0'), 'train_loss_bbox': tensor(0.0119, device='cuda:0'), 'train_loss_giou': tensor(0.0696, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4463, device='cuda:0'), 'validation_loss_ce': tensor(0.3245, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0437, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4463, device='cuda:0'), 'train_loss_ce': tensor(0.3245, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0437, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4833, device='cuda:0'), 'validation_loss_ce': tensor(0.3235, device='cuda:0'), 'validation_loss_bbox': tensor(0.0080, device='cuda:0'), 'validation_loss_giou': tensor(0.0599, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4833, device='cuda:0'), 'train_loss_ce': tensor(0.3235, device='cuda:0'), 'train_loss_bbox': tensor(0.0080, device='cuda:0'), 'train_loss_giou': tensor(0.0599, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4979, device='cuda:0'), 'validation_loss_ce': tensor(0.3226, device='cuda:0'), 'validation_loss_bbox': tensor(0.0095, device='cuda:0'), 'validation_loss_giou': tensor(0.0640, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4979, device='cuda:0'), 'train_loss_ce': tensor(0.3226, device='cuda:0'), 'train_loss_bbox': tensor(0.0095, device='cuda:0'), 'train_loss_giou': tensor(0.0640, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4111, device='cuda:0'), 'validation_loss_ce': tensor(0.3220, device='cuda:0'), 'validation_loss_bbox': tensor(0.0056, device='cuda:0'), 'validation_loss_giou': tensor(0.0305, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4111, device='cuda:0'), 'train_loss_ce': tensor(0.3220, device='cuda:0'), 'train_loss_bbox': tensor(0.0056, device='cuda:0'), 'train_loss_giou': tensor(0.0305, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4558, device='cuda:0'), 'validation_loss_ce': tensor(0.3214, device='cuda:0'), 'validation_loss_bbox': tensor(0.0060, device='cuda:0'), 'validation_loss_giou': tensor(0.0523, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4558, device='cuda:0'), 'train_loss_ce': tensor(0.3214, device='cuda:0'), 'train_loss_bbox': tensor(0.0060, device='cuda:0'), 'train_loss_giou': tensor(0.0523, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4333, device='cuda:0'), 'validation_loss_ce': tensor(0.3203, device='cuda:0'), 'validation_loss_bbox': tensor(0.0060, device='cuda:0'), 'validation_loss_giou': tensor(0.0416, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4333, device='cuda:0'), 'train_loss_ce': tensor(0.3203, device='cuda:0'), 'train_loss_bbox': tensor(0.0060, device='cuda:0'), 'train_loss_giou': tensor(0.0416, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4744, device='cuda:0'), 'validation_loss_ce': tensor(0.3192, device='cuda:0'), 'validation_loss_bbox': tensor(0.0080, device='cuda:0'), 'validation_loss_giou': tensor(0.0575, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4744, device='cuda:0'), 'train_loss_ce': tensor(0.3192, device='cuda:0'), 'train_loss_bbox': tensor(0.0080, device='cuda:0'), 'train_loss_giou': tensor(0.0575, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4671, device='cuda:0'), 'validation_loss_ce': tensor(0.3184, device='cuda:0'), 'validation_loss_bbox': tensor(0.0082, device='cuda:0'), 'validation_loss_giou': tensor(0.0539, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4671, device='cuda:0'), 'train_loss_ce': tensor(0.3184, device='cuda:0'), 'train_loss_bbox': tensor(0.0082, device='cuda:0'), 'train_loss_giou': tensor(0.0539, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4212, device='cuda:0'), 'validation_loss_ce': tensor(0.3180, device='cuda:0'), 'validation_loss_bbox': tensor(0.0059, device='cuda:0'), 'validation_loss_giou': tensor(0.0369, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4212, device='cuda:0'), 'train_loss_ce': tensor(0.3180, device='cuda:0'), 'train_loss_bbox': tensor(0.0059, device='cuda:0'), 'train_loss_giou': tensor(0.0369, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4802, device='cuda:0'), 'validation_loss_ce': tensor(0.3176, device='cuda:0'), 'validation_loss_bbox': tensor(0.0088, device='cuda:0'), 'validation_loss_giou': tensor(0.0592, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4802, device='cuda:0'), 'train_loss_ce': tensor(0.3176, device='cuda:0'), 'train_loss_bbox': tensor(0.0088, device='cuda:0'), 'train_loss_giou': tensor(0.0592, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4904, device='cuda:0'), 'validation_loss_ce': tensor(0.3187, device='cuda:0'), 'validation_loss_bbox': tensor(0.0081, device='cuda:0'), 'validation_loss_giou': tensor(0.0656, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4904, device='cuda:0'), 'train_loss_ce': tensor(0.3187, device='cuda:0'), 'train_loss_bbox': tensor(0.0081, device='cuda:0'), 'train_loss_giou': tensor(0.0656, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4515, device='cuda:0'), 'validation_loss_ce': tensor(0.3179, device='cuda:0'), 'validation_loss_bbox': tensor(0.0068, device='cuda:0'), 'validation_loss_giou': tensor(0.0499, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4515, device='cuda:0'), 'train_loss_ce': tensor(0.3179, device='cuda:0'), 'train_loss_bbox': tensor(0.0068, device='cuda:0'), 'train_loss_giou': tensor(0.0499, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4780, device='cuda:0'), 'validation_loss_ce': tensor(0.3167, device='cuda:0'), 'validation_loss_bbox': tensor(0.0093, device='cuda:0'), 'validation_loss_giou': tensor(0.0575, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4780, device='cuda:0'), 'train_loss_ce': tensor(0.3167, device='cuda:0'), 'train_loss_bbox': tensor(0.0093, device='cuda:0'), 'train_loss_giou': tensor(0.0575, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4906, device='cuda:0'), 'validation_loss_ce': tensor(0.3158, device='cuda:0'), 'validation_loss_bbox': tensor(0.0087, device='cuda:0'), 'validation_loss_giou': tensor(0.0657, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4906, device='cuda:0'), 'train_loss_ce': tensor(0.3158, device='cuda:0'), 'train_loss_bbox': tensor(0.0087, device='cuda:0'), 'train_loss_giou': tensor(0.0657, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4108, device='cuda:0'), 'validation_loss_ce': tensor(0.3151, device='cuda:0'), 'validation_loss_bbox': tensor(0.0066, device='cuda:0'), 'validation_loss_giou': tensor(0.0314, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4108, device='cuda:0'), 'train_loss_ce': tensor(0.3151, device='cuda:0'), 'train_loss_bbox': tensor(0.0066, device='cuda:0'), 'train_loss_giou': tensor(0.0314, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4721, device='cuda:0'), 'validation_loss_ce': tensor(0.3146, device='cuda:0'), 'validation_loss_bbox': tensor(0.0072, device='cuda:0'), 'validation_loss_giou': tensor(0.0608, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4721, device='cuda:0'), 'train_loss_ce': tensor(0.3146, device='cuda:0'), 'train_loss_bbox': tensor(0.0072, device='cuda:0'), 'train_loss_giou': tensor(0.0608, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5061, device='cuda:0'), 'validation_loss_ce': tensor(0.3141, device='cuda:0'), 'validation_loss_bbox': tensor(0.0114, device='cuda:0'), 'validation_loss_giou': tensor(0.0674, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5061, device='cuda:0'), 'train_loss_ce': tensor(0.3141, device='cuda:0'), 'train_loss_bbox': tensor(0.0114, device='cuda:0'), 'train_loss_giou': tensor(0.0674, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4657, device='cuda:0'), 'validation_loss_ce': tensor(0.3133, device='cuda:0'), 'validation_loss_bbox': tensor(0.0102, device='cuda:0'), 'validation_loss_giou': tensor(0.0506, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4657, device='cuda:0'), 'train_loss_ce': tensor(0.3133, device='cuda:0'), 'train_loss_bbox': tensor(0.0102, device='cuda:0'), 'train_loss_giou': tensor(0.0506, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5113, device='cuda:0'), 'validation_loss_ce': tensor(0.3121, device='cuda:0'), 'validation_loss_bbox': tensor(0.0095, device='cuda:0'), 'validation_loss_giou': tensor(0.0758, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5113, device='cuda:0'), 'train_loss_ce': tensor(0.3121, device='cuda:0'), 'train_loss_bbox': tensor(0.0095, device='cuda:0'), 'train_loss_giou': tensor(0.0758, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5286, device='cuda:0'), 'validation_loss_ce': tensor(0.3128, device='cuda:0'), 'validation_loss_bbox': tensor(0.0131, device='cuda:0'), 'validation_loss_giou': tensor(0.0752, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5286, device='cuda:0'), 'train_loss_ce': tensor(0.3128, device='cuda:0'), 'train_loss_bbox': tensor(0.0131, device='cuda:0'), 'train_loss_giou': tensor(0.0752, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4768, device='cuda:0'), 'validation_loss_ce': tensor(0.3118, device='cuda:0'), 'validation_loss_bbox': tensor(0.0095, device='cuda:0'), 'validation_loss_giou': tensor(0.0587, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4768, device='cuda:0'), 'train_loss_ce': tensor(0.3118, device='cuda:0'), 'train_loss_bbox': tensor(0.0095, device='cuda:0'), 'train_loss_giou': tensor(0.0587, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4146, device='cuda:0'), 'validation_loss_ce': tensor(0.3109, device='cuda:0'), 'validation_loss_bbox': tensor(0.0059, device='cuda:0'), 'validation_loss_giou': tensor(0.0371, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4146, device='cuda:0'), 'train_loss_ce': tensor(0.3109, device='cuda:0'), 'train_loss_bbox': tensor(0.0059, device='cuda:0'), 'train_loss_giou': tensor(0.0371, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4642, device='cuda:0'), 'validation_loss_ce': tensor(0.3103, device='cuda:0'), 'validation_loss_bbox': tensor(0.0101, device='cuda:0'), 'validation_loss_giou': tensor(0.0517, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4642, device='cuda:0'), 'train_loss_ce': tensor(0.3103, device='cuda:0'), 'train_loss_bbox': tensor(0.0101, device='cuda:0'), 'train_loss_giou': tensor(0.0517, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4811, device='cuda:0'), 'validation_loss_ce': tensor(0.3096, device='cuda:0'), 'validation_loss_bbox': tensor(0.0108, device='cuda:0'), 'validation_loss_giou': tensor(0.0588, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4811, device='cuda:0'), 'train_loss_ce': tensor(0.3096, device='cuda:0'), 'train_loss_bbox': tensor(0.0108, device='cuda:0'), 'train_loss_giou': tensor(0.0588, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4223, device='cuda:0'), 'validation_loss_ce': tensor(0.3086, device='cuda:0'), 'validation_loss_bbox': tensor(0.0071, device='cuda:0'), 'validation_loss_giou': tensor(0.0390, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4223, device='cuda:0'), 'train_loss_ce': tensor(0.3086, device='cuda:0'), 'train_loss_bbox': tensor(0.0071, device='cuda:0'), 'train_loss_giou': tensor(0.0390, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4187, device='cuda:0'), 'validation_loss_ce': tensor(0.3080, device='cuda:0'), 'validation_loss_bbox': tensor(0.0067, device='cuda:0'), 'validation_loss_giou': tensor(0.0386, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4187, device='cuda:0'), 'train_loss_ce': tensor(0.3080, device='cuda:0'), 'train_loss_bbox': tensor(0.0067, device='cuda:0'), 'train_loss_giou': tensor(0.0386, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4132, device='cuda:0'), 'validation_loss_ce': tensor(0.3078, device='cuda:0'), 'validation_loss_bbox': tensor(0.0070, device='cuda:0'), 'validation_loss_giou': tensor(0.0353, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4132, device='cuda:0'), 'train_loss_ce': tensor(0.3078, device='cuda:0'), 'train_loss_bbox': tensor(0.0070, device='cuda:0'), 'train_loss_giou': tensor(0.0353, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4466, device='cuda:0'), 'validation_loss_ce': tensor(0.3072, device='cuda:0'), 'validation_loss_bbox': tensor(0.0084, device='cuda:0'), 'validation_loss_giou': tensor(0.0487, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4466, device='cuda:0'), 'train_loss_ce': tensor(0.3072, device='cuda:0'), 'train_loss_bbox': tensor(0.0084, device='cuda:0'), 'train_loss_giou': tensor(0.0487, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4490, device='cuda:0'), 'validation_loss_ce': tensor(0.3064, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0489, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4490, device='cuda:0'), 'train_loss_ce': tensor(0.3064, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0489, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4229, device='cuda:0'), 'validation_loss_ce': tensor(0.3054, device='cuda:0'), 'validation_loss_bbox': tensor(0.0076, device='cuda:0'), 'validation_loss_giou': tensor(0.0396, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4229, device='cuda:0'), 'train_loss_ce': tensor(0.3054, device='cuda:0'), 'train_loss_bbox': tensor(0.0076, device='cuda:0'), 'train_loss_giou': tensor(0.0396, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4345, device='cuda:0'), 'validation_loss_ce': tensor(0.3045, device='cuda:0'), 'validation_loss_bbox': tensor(0.0070, device='cuda:0'), 'validation_loss_giou': tensor(0.0475, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4345, device='cuda:0'), 'train_loss_ce': tensor(0.3045, device='cuda:0'), 'train_loss_bbox': tensor(0.0070, device='cuda:0'), 'train_loss_giou': tensor(0.0475, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4229, device='cuda:0'), 'validation_loss_ce': tensor(0.3037, device='cuda:0'), 'validation_loss_bbox': tensor(0.0072, device='cuda:0'), 'validation_loss_giou': tensor(0.0416, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4229, device='cuda:0'), 'train_loss_ce': tensor(0.3037, device='cuda:0'), 'train_loss_bbox': tensor(0.0072, device='cuda:0'), 'train_loss_giou': tensor(0.0416, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4135, device='cuda:0'), 'validation_loss_ce': tensor(0.3026, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0382, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4135, device='cuda:0'), 'train_loss_ce': tensor(0.3026, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0382, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4305, device='cuda:0'), 'validation_loss_ce': tensor(0.3016, device='cuda:0'), 'validation_loss_bbox': tensor(0.0076, device='cuda:0'), 'validation_loss_giou': tensor(0.0455, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4305, device='cuda:0'), 'train_loss_ce': tensor(0.3016, device='cuda:0'), 'train_loss_bbox': tensor(0.0076, device='cuda:0'), 'train_loss_giou': tensor(0.0455, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4266, device='cuda:0'), 'validation_loss_ce': tensor(0.3009, device='cuda:0'), 'validation_loss_bbox': tensor(0.0060, device='cuda:0'), 'validation_loss_giou': tensor(0.0477, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4266, device='cuda:0'), 'train_loss_ce': tensor(0.3009, device='cuda:0'), 'train_loss_bbox': tensor(0.0060, device='cuda:0'), 'train_loss_giou': tensor(0.0477, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4173, device='cuda:0'), 'validation_loss_ce': tensor(0.3001, device='cuda:0'), 'validation_loss_bbox': tensor(0.0060, device='cuda:0'), 'validation_loss_giou': tensor(0.0437, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4173, device='cuda:0'), 'train_loss_ce': tensor(0.3001, device='cuda:0'), 'train_loss_bbox': tensor(0.0060, device='cuda:0'), 'train_loss_giou': tensor(0.0437, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4494, device='cuda:0'), 'validation_loss_ce': tensor(0.2989, device='cuda:0'), 'validation_loss_bbox': tensor(0.0081, device='cuda:0'), 'validation_loss_giou': tensor(0.0550, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4494, device='cuda:0'), 'train_loss_ce': tensor(0.2989, device='cuda:0'), 'train_loss_bbox': tensor(0.0081, device='cuda:0'), 'train_loss_giou': tensor(0.0550, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4788, device='cuda:0'), 'validation_loss_ce': tensor(0.2980, device='cuda:0'), 'validation_loss_bbox': tensor(0.0110, device='cuda:0'), 'validation_loss_giou': tensor(0.0628, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4788, device='cuda:0'), 'train_loss_ce': tensor(0.2980, device='cuda:0'), 'train_loss_bbox': tensor(0.0110, device='cuda:0'), 'train_loss_giou': tensor(0.0628, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4453, device='cuda:0'), 'validation_loss_ce': tensor(0.2975, device='cuda:0'), 'validation_loss_bbox': tensor(0.0085, device='cuda:0'), 'validation_loss_giou': tensor(0.0527, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4453, device='cuda:0'), 'train_loss_ce': tensor(0.2975, device='cuda:0'), 'train_loss_bbox': tensor(0.0085, device='cuda:0'), 'train_loss_giou': tensor(0.0527, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4510, device='cuda:0'), 'validation_loss_ce': tensor(0.2969, device='cuda:0'), 'validation_loss_bbox': tensor(0.0082, device='cuda:0'), 'validation_loss_giou': tensor(0.0565, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4510, device='cuda:0'), 'train_loss_ce': tensor(0.2969, device='cuda:0'), 'train_loss_bbox': tensor(0.0082, device='cuda:0'), 'train_loss_giou': tensor(0.0565, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4314, device='cuda:0'), 'validation_loss_ce': tensor(0.2960, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0453, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4314, device='cuda:0'), 'train_loss_ce': tensor(0.2960, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0453, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4589, device='cuda:0'), 'validation_loss_ce': tensor(0.2949, device='cuda:0'), 'validation_loss_bbox': tensor(0.0079, device='cuda:0'), 'validation_loss_giou': tensor(0.0621, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4589, device='cuda:0'), 'train_loss_ce': tensor(0.2949, device='cuda:0'), 'train_loss_bbox': tensor(0.0079, device='cuda:0'), 'train_loss_giou': tensor(0.0621, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4830, device='cuda:0'), 'validation_loss_ce': tensor(0.2938, device='cuda:0'), 'validation_loss_bbox': tensor(0.0081, device='cuda:0'), 'validation_loss_giou': tensor(0.0744, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4830, device='cuda:0'), 'train_loss_ce': tensor(0.2938, device='cuda:0'), 'train_loss_bbox': tensor(0.0081, device='cuda:0'), 'train_loss_giou': tensor(0.0744, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4463, device='cuda:0'), 'validation_loss_ce': tensor(0.2932, device='cuda:0'), 'validation_loss_bbox': tensor(0.0082, device='cuda:0'), 'validation_loss_giou': tensor(0.0561, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4463, device='cuda:0'), 'train_loss_ce': tensor(0.2932, device='cuda:0'), 'train_loss_bbox': tensor(0.0082, device='cuda:0'), 'train_loss_giou': tensor(0.0561, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4154, device='cuda:0'), 'validation_loss_ce': tensor(0.2929, device='cuda:0'), 'validation_loss_bbox': tensor(0.0074, device='cuda:0'), 'validation_loss_giou': tensor(0.0427, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4154, device='cuda:0'), 'train_loss_ce': tensor(0.2929, device='cuda:0'), 'train_loss_bbox': tensor(0.0074, device='cuda:0'), 'train_loss_giou': tensor(0.0427, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4529, device='cuda:0'), 'validation_loss_ce': tensor(0.2923, device='cuda:0'), 'validation_loss_bbox': tensor(0.0103, device='cuda:0'), 'validation_loss_giou': tensor(0.0547, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4529, device='cuda:0'), 'train_loss_ce': tensor(0.2923, device='cuda:0'), 'train_loss_bbox': tensor(0.0103, device='cuda:0'), 'train_loss_giou': tensor(0.0547, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4542, device='cuda:0'), 'validation_loss_ce': tensor(0.2918, device='cuda:0'), 'validation_loss_bbox': tensor(0.0100, device='cuda:0'), 'validation_loss_giou': tensor(0.0561, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4542, device='cuda:0'), 'train_loss_ce': tensor(0.2918, device='cuda:0'), 'train_loss_bbox': tensor(0.0100, device='cuda:0'), 'train_loss_giou': tensor(0.0561, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4192, device='cuda:0'), 'validation_loss_ce': tensor(0.2914, device='cuda:0'), 'validation_loss_bbox': tensor(0.0080, device='cuda:0'), 'validation_loss_giou': tensor(0.0439, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4192, device='cuda:0'), 'train_loss_ce': tensor(0.2914, device='cuda:0'), 'train_loss_bbox': tensor(0.0080, device='cuda:0'), 'train_loss_giou': tensor(0.0439, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4113, device='cuda:0'), 'validation_loss_ce': tensor(0.2903, device='cuda:0'), 'validation_loss_bbox': tensor(0.0066, device='cuda:0'), 'validation_loss_giou': tensor(0.0441, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4113, device='cuda:0'), 'train_loss_ce': tensor(0.2903, device='cuda:0'), 'train_loss_bbox': tensor(0.0066, device='cuda:0'), 'train_loss_giou': tensor(0.0441, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4022, device='cuda:0'), 'validation_loss_ce': tensor(0.2897, device='cuda:0'), 'validation_loss_bbox': tensor(0.0076, device='cuda:0'), 'validation_loss_giou': tensor(0.0372, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4022, device='cuda:0'), 'train_loss_ce': tensor(0.2897, device='cuda:0'), 'train_loss_bbox': tensor(0.0076, device='cuda:0'), 'train_loss_giou': tensor(0.0372, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4346, device='cuda:0'), 'validation_loss_ce': tensor(0.2879, device='cuda:0'), 'validation_loss_bbox': tensor(0.0107, device='cuda:0'), 'validation_loss_giou': tensor(0.0467, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4346, device='cuda:0'), 'train_loss_ce': tensor(0.2879, device='cuda:0'), 'train_loss_bbox': tensor(0.0107, device='cuda:0'), 'train_loss_giou': tensor(0.0467, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4227, device='cuda:0'), 'validation_loss_ce': tensor(0.2864, device='cuda:0'), 'validation_loss_bbox': tensor(0.0102, device='cuda:0'), 'validation_loss_giou': tensor(0.0426, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4227, device='cuda:0'), 'train_loss_ce': tensor(0.2864, device='cuda:0'), 'train_loss_bbox': tensor(0.0102, device='cuda:0'), 'train_loss_giou': tensor(0.0426, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3907, device='cuda:0'), 'validation_loss_ce': tensor(0.2854, device='cuda:0'), 'validation_loss_bbox': tensor(0.0071, device='cuda:0'), 'validation_loss_giou': tensor(0.0349, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3907, device='cuda:0'), 'train_loss_ce': tensor(0.2854, device='cuda:0'), 'train_loss_bbox': tensor(0.0071, device='cuda:0'), 'train_loss_giou': tensor(0.0349, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4241, device='cuda:0'), 'validation_loss_ce': tensor(0.2848, device='cuda:0'), 'validation_loss_bbox': tensor(0.0070, device='cuda:0'), 'validation_loss_giou': tensor(0.0522, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4241, device='cuda:0'), 'train_loss_ce': tensor(0.2848, device='cuda:0'), 'train_loss_bbox': tensor(0.0070, device='cuda:0'), 'train_loss_giou': tensor(0.0522, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4297, device='cuda:0'), 'validation_loss_ce': tensor(0.2841, device='cuda:0'), 'validation_loss_bbox': tensor(0.0083, device='cuda:0'), 'validation_loss_giou': tensor(0.0520, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4297, device='cuda:0'), 'train_loss_ce': tensor(0.2841, device='cuda:0'), 'train_loss_bbox': tensor(0.0083, device='cuda:0'), 'train_loss_giou': tensor(0.0520, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4233, device='cuda:0'), 'validation_loss_ce': tensor(0.2831, device='cuda:0'), 'validation_loss_bbox': tensor(0.0082, device='cuda:0'), 'validation_loss_giou': tensor(0.0497, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4233, device='cuda:0'), 'train_loss_ce': tensor(0.2831, device='cuda:0'), 'train_loss_bbox': tensor(0.0082, device='cuda:0'), 'train_loss_giou': tensor(0.0497, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4607, device='cuda:0'), 'validation_loss_ce': tensor(0.2818, device='cuda:0'), 'validation_loss_bbox': tensor(0.0105, device='cuda:0'), 'validation_loss_giou': tensor(0.0631, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4607, device='cuda:0'), 'train_loss_ce': tensor(0.2818, device='cuda:0'), 'train_loss_bbox': tensor(0.0105, device='cuda:0'), 'train_loss_giou': tensor(0.0631, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4320, device='cuda:0'), 'validation_loss_ce': tensor(0.2806, device='cuda:0'), 'validation_loss_bbox': tensor(0.0125, device='cuda:0'), 'validation_loss_giou': tensor(0.0444, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4320, device='cuda:0'), 'train_loss_ce': tensor(0.2806, device='cuda:0'), 'train_loss_bbox': tensor(0.0125, device='cuda:0'), 'train_loss_giou': tensor(0.0444, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5268, device='cuda:0'), 'validation_loss_ce': tensor(0.2799, device='cuda:0'), 'validation_loss_bbox': tensor(0.0153, device='cuda:0'), 'validation_loss_giou': tensor(0.0852, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5268, device='cuda:0'), 'train_loss_ce': tensor(0.2799, device='cuda:0'), 'train_loss_bbox': tensor(0.0153, device='cuda:0'), 'train_loss_giou': tensor(0.0852, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4750, device='cuda:0'), 'validation_loss_ce': tensor(0.2792, device='cuda:0'), 'validation_loss_bbox': tensor(0.0100, device='cuda:0'), 'validation_loss_giou': tensor(0.0729, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4750, device='cuda:0'), 'train_loss_ce': tensor(0.2792, device='cuda:0'), 'train_loss_bbox': tensor(0.0100, device='cuda:0'), 'train_loss_giou': tensor(0.0729, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4019, device='cuda:0'), 'validation_loss_ce': tensor(0.2797, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0438, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4019, device='cuda:0'), 'train_loss_ce': tensor(0.2797, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0438, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5179, device='cuda:0'), 'validation_loss_ce': tensor(0.2787, device='cuda:0'), 'validation_loss_bbox': tensor(0.0135, device='cuda:0'), 'validation_loss_giou': tensor(0.0857, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5179, device='cuda:0'), 'train_loss_ce': tensor(0.2787, device='cuda:0'), 'train_loss_bbox': tensor(0.0135, device='cuda:0'), 'train_loss_giou': tensor(0.0857, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5993, device='cuda:0'), 'validation_loss_ce': tensor(0.2791, device='cuda:0'), 'validation_loss_bbox': tensor(0.0177, device='cuda:0'), 'validation_loss_giou': tensor(0.1159, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5993, device='cuda:0'), 'train_loss_ce': tensor(0.2791, device='cuda:0'), 'train_loss_bbox': tensor(0.0177, device='cuda:0'), 'train_loss_giou': tensor(0.1159, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5874, device='cuda:0'), 'validation_loss_ce': tensor(0.2777, device='cuda:0'), 'validation_loss_bbox': tensor(0.0164, device='cuda:0'), 'validation_loss_giou': tensor(0.1137, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5874, device='cuda:0'), 'train_loss_ce': tensor(0.2777, device='cuda:0'), 'train_loss_bbox': tensor(0.0164, device='cuda:0'), 'train_loss_giou': tensor(0.1137, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5082, device='cuda:0'), 'validation_loss_ce': tensor(0.2767, device='cuda:0'), 'validation_loss_bbox': tensor(0.0121, device='cuda:0'), 'validation_loss_giou': tensor(0.0854, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5082, device='cuda:0'), 'train_loss_ce': tensor(0.2767, device='cuda:0'), 'train_loss_bbox': tensor(0.0121, device='cuda:0'), 'train_loss_giou': tensor(0.0854, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4779, device='cuda:0'), 'validation_loss_ce': tensor(0.2739, device='cuda:0'), 'validation_loss_bbox': tensor(0.0120, device='cuda:0'), 'validation_loss_giou': tensor(0.0719, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4779, device='cuda:0'), 'train_loss_ce': tensor(0.2739, device='cuda:0'), 'train_loss_bbox': tensor(0.0120, device='cuda:0'), 'train_loss_giou': tensor(0.0719, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4743, device='cuda:0'), 'validation_loss_ce': tensor(0.2737, device='cuda:0'), 'validation_loss_bbox': tensor(0.0114, device='cuda:0'), 'validation_loss_giou': tensor(0.0718, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4743, device='cuda:0'), 'train_loss_ce': tensor(0.2737, device='cuda:0'), 'train_loss_bbox': tensor(0.0114, device='cuda:0'), 'train_loss_giou': tensor(0.0718, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.5339, device='cuda:0'), 'validation_loss_ce': tensor(0.2736, device='cuda:0'), 'validation_loss_bbox': tensor(0.0139, device='cuda:0'), 'validation_loss_giou': tensor(0.0954, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.5339, device='cuda:0'), 'train_loss_ce': tensor(0.2736, device='cuda:0'), 'train_loss_bbox': tensor(0.0139, device='cuda:0'), 'train_loss_giou': tensor(0.0954, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4771, device='cuda:0'), 'validation_loss_ce': tensor(0.2730, device='cuda:0'), 'validation_loss_bbox': tensor(0.0117, device='cuda:0'), 'validation_loss_giou': tensor(0.0727, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4771, device='cuda:0'), 'train_loss_ce': tensor(0.2730, device='cuda:0'), 'train_loss_bbox': tensor(0.0117, device='cuda:0'), 'train_loss_giou': tensor(0.0727, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4488, device='cuda:0'), 'validation_loss_ce': tensor(0.2720, device='cuda:0'), 'validation_loss_bbox': tensor(0.0107, device='cuda:0'), 'validation_loss_giou': tensor(0.0617, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4488, device='cuda:0'), 'train_loss_ce': tensor(0.2720, device='cuda:0'), 'train_loss_bbox': tensor(0.0107, device='cuda:0'), 'train_loss_giou': tensor(0.0617, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4195, device='cuda:0'), 'validation_loss_ce': tensor(0.2710, device='cuda:0'), 'validation_loss_bbox': tensor(0.0086, device='cuda:0'), 'validation_loss_giou': tensor(0.0528, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4195, device='cuda:0'), 'train_loss_ce': tensor(0.2710, device='cuda:0'), 'train_loss_bbox': tensor(0.0086, device='cuda:0'), 'train_loss_giou': tensor(0.0528, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4454, device='cuda:0'), 'validation_loss_ce': tensor(0.2701, device='cuda:0'), 'validation_loss_bbox': tensor(0.0110, device='cuda:0'), 'validation_loss_giou': tensor(0.0603, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4454, device='cuda:0'), 'train_loss_ce': tensor(0.2701, device='cuda:0'), 'train_loss_bbox': tensor(0.0110, device='cuda:0'), 'train_loss_giou': tensor(0.0603, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4524, device='cuda:0'), 'validation_loss_ce': tensor(0.2698, device='cuda:0'), 'validation_loss_bbox': tensor(0.0131, device='cuda:0'), 'validation_loss_giou': tensor(0.0585, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4524, device='cuda:0'), 'train_loss_ce': tensor(0.2698, device='cuda:0'), 'train_loss_bbox': tensor(0.0131, device='cuda:0'), 'train_loss_giou': tensor(0.0585, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4416, device='cuda:0'), 'validation_loss_ce': tensor(0.2695, device='cuda:0'), 'validation_loss_bbox': tensor(0.0112, device='cuda:0'), 'validation_loss_giou': tensor(0.0581, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4416, device='cuda:0'), 'train_loss_ce': tensor(0.2695, device='cuda:0'), 'train_loss_bbox': tensor(0.0112, device='cuda:0'), 'train_loss_giou': tensor(0.0581, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3454, device='cuda:0'), 'validation_loss_ce': tensor(0.2692, device='cuda:0'), 'validation_loss_bbox': tensor(0.0054, device='cuda:0'), 'validation_loss_giou': tensor(0.0246, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3454, device='cuda:0'), 'train_loss_ce': tensor(0.2692, device='cuda:0'), 'train_loss_bbox': tensor(0.0054, device='cuda:0'), 'train_loss_giou': tensor(0.0246, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4792, device='cuda:0'), 'validation_loss_ce': tensor(0.2688, device='cuda:0'), 'validation_loss_bbox': tensor(0.0134, device='cuda:0'), 'validation_loss_giou': tensor(0.0717, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4792, device='cuda:0'), 'train_loss_ce': tensor(0.2688, device='cuda:0'), 'train_loss_bbox': tensor(0.0134, device='cuda:0'), 'train_loss_giou': tensor(0.0717, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4578, device='cuda:0'), 'validation_loss_ce': tensor(0.2683, device='cuda:0'), 'validation_loss_bbox': tensor(0.0135, device='cuda:0'), 'validation_loss_giou': tensor(0.0610, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4578, device='cuda:0'), 'train_loss_ce': tensor(0.2683, device='cuda:0'), 'train_loss_bbox': tensor(0.0135, device='cuda:0'), 'train_loss_giou': tensor(0.0610, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4862, device='cuda:0'), 'validation_loss_ce': tensor(0.2675, device='cuda:0'), 'validation_loss_bbox': tensor(0.0107, device='cuda:0'), 'validation_loss_giou': tensor(0.0827, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4862, device='cuda:0'), 'train_loss_ce': tensor(0.2675, device='cuda:0'), 'train_loss_bbox': tensor(0.0107, device='cuda:0'), 'train_loss_giou': tensor(0.0827, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4948, device='cuda:0'), 'validation_loss_ce': tensor(0.2653, device='cuda:0'), 'validation_loss_bbox': tensor(0.0128, device='cuda:0'), 'validation_loss_giou': tensor(0.0829, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4948, device='cuda:0'), 'train_loss_ce': tensor(0.2653, device='cuda:0'), 'train_loss_bbox': tensor(0.0128, device='cuda:0'), 'train_loss_giou': tensor(0.0829, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4852, device='cuda:0'), 'validation_loss_ce': tensor(0.2645, device='cuda:0'), 'validation_loss_bbox': tensor(0.0147, device='cuda:0'), 'validation_loss_giou': tensor(0.0736, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4852, device='cuda:0'), 'train_loss_ce': tensor(0.2645, device='cuda:0'), 'train_loss_bbox': tensor(0.0147, device='cuda:0'), 'train_loss_giou': tensor(0.0736, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4610, device='cuda:0'), 'validation_loss_ce': tensor(0.2641, device='cuda:0'), 'validation_loss_bbox': tensor(0.0124, device='cuda:0'), 'validation_loss_giou': tensor(0.0675, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4610, device='cuda:0'), 'train_loss_ce': tensor(0.2641, device='cuda:0'), 'train_loss_bbox': tensor(0.0124, device='cuda:0'), 'train_loss_giou': tensor(0.0675, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4039, device='cuda:0'), 'validation_loss_ce': tensor(0.2637, device='cuda:0'), 'validation_loss_bbox': tensor(0.0065, device='cuda:0'), 'validation_loss_giou': tensor(0.0540, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4039, device='cuda:0'), 'train_loss_ce': tensor(0.2637, device='cuda:0'), 'train_loss_bbox': tensor(0.0065, device='cuda:0'), 'train_loss_giou': tensor(0.0540, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4428, device='cuda:0'), 'validation_loss_ce': tensor(0.2628, device='cuda:0'), 'validation_loss_bbox': tensor(0.0092, device='cuda:0'), 'validation_loss_giou': tensor(0.0670, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4428, device='cuda:0'), 'train_loss_ce': tensor(0.2628, device='cuda:0'), 'train_loss_bbox': tensor(0.0092, device='cuda:0'), 'train_loss_giou': tensor(0.0670, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4629, device='cuda:0'), 'validation_loss_ce': tensor(0.2618, device='cuda:0'), 'validation_loss_bbox': tensor(0.0115, device='cuda:0'), 'validation_loss_giou': tensor(0.0717, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4629, device='cuda:0'), 'train_loss_ce': tensor(0.2618, device='cuda:0'), 'train_loss_bbox': tensor(0.0115, device='cuda:0'), 'train_loss_giou': tensor(0.0717, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4212, device='cuda:0'), 'validation_loss_ce': tensor(0.2607, device='cuda:0'), 'validation_loss_bbox': tensor(0.0090, device='cuda:0'), 'validation_loss_giou': tensor(0.0577, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4212, device='cuda:0'), 'train_loss_ce': tensor(0.2607, device='cuda:0'), 'train_loss_bbox': tensor(0.0090, device='cuda:0'), 'train_loss_giou': tensor(0.0577, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4288, device='cuda:0'), 'validation_loss_ce': tensor(0.2595, device='cuda:0'), 'validation_loss_bbox': tensor(0.0077, device='cuda:0'), 'validation_loss_giou': tensor(0.0654, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4288, device='cuda:0'), 'train_loss_ce': tensor(0.2595, device='cuda:0'), 'train_loss_bbox': tensor(0.0077, device='cuda:0'), 'train_loss_giou': tensor(0.0654, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4233, device='cuda:0'), 'validation_loss_ce': tensor(0.2584, device='cuda:0'), 'validation_loss_bbox': tensor(0.0086, device='cuda:0'), 'validation_loss_giou': tensor(0.0610, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4233, device='cuda:0'), 'train_loss_ce': tensor(0.2584, device='cuda:0'), 'train_loss_bbox': tensor(0.0086, device='cuda:0'), 'train_loss_giou': tensor(0.0610, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3875, device='cuda:0'), 'validation_loss_ce': tensor(0.2576, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0478, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3875, device='cuda:0'), 'train_loss_ce': tensor(0.2576, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0478, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3998, device='cuda:0'), 'validation_loss_ce': tensor(0.2575, device='cuda:0'), 'validation_loss_bbox': tensor(0.0054, device='cuda:0'), 'validation_loss_giou': tensor(0.0577, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3998, device='cuda:0'), 'train_loss_ce': tensor(0.2575, device='cuda:0'), 'train_loss_bbox': tensor(0.0054, device='cuda:0'), 'train_loss_giou': tensor(0.0577, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3661, device='cuda:0'), 'validation_loss_ce': tensor(0.2563, device='cuda:0'), 'validation_loss_bbox': tensor(0.0063, device='cuda:0'), 'validation_loss_giou': tensor(0.0391, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3661, device='cuda:0'), 'train_loss_ce': tensor(0.2563, device='cuda:0'), 'train_loss_bbox': tensor(0.0063, device='cuda:0'), 'train_loss_giou': tensor(0.0391, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4554, device='cuda:0'), 'validation_loss_ce': tensor(0.2554, device='cuda:0'), 'validation_loss_bbox': tensor(0.0079, device='cuda:0'), 'validation_loss_giou': tensor(0.0804, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4554, device='cuda:0'), 'train_loss_ce': tensor(0.2554, device='cuda:0'), 'train_loss_bbox': tensor(0.0079, device='cuda:0'), 'train_loss_giou': tensor(0.0804, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4649, device='cuda:0'), 'validation_loss_ce': tensor(0.2546, device='cuda:0'), 'validation_loss_bbox': tensor(0.0063, device='cuda:0'), 'validation_loss_giou': tensor(0.0894, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4649, device='cuda:0'), 'train_loss_ce': tensor(0.2546, device='cuda:0'), 'train_loss_bbox': tensor(0.0063, device='cuda:0'), 'train_loss_giou': tensor(0.0894, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4299, device='cuda:0'), 'validation_loss_ce': tensor(0.2539, device='cuda:0'), 'validation_loss_bbox': tensor(0.0071, device='cuda:0'), 'validation_loss_giou': tensor(0.0702, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4299, device='cuda:0'), 'train_loss_ce': tensor(0.2539, device='cuda:0'), 'train_loss_bbox': tensor(0.0071, device='cuda:0'), 'train_loss_giou': tensor(0.0702, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3893, device='cuda:0'), 'validation_loss_ce': tensor(0.2535, device='cuda:0'), 'validation_loss_bbox': tensor(0.0070, device='cuda:0'), 'validation_loss_giou': tensor(0.0504, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3893, device='cuda:0'), 'train_loss_ce': tensor(0.2535, device='cuda:0'), 'train_loss_bbox': tensor(0.0070, device='cuda:0'), 'train_loss_giou': tensor(0.0504, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3641, device='cuda:0'), 'validation_loss_ce': tensor(0.2529, device='cuda:0'), 'validation_loss_bbox': tensor(0.0059, device='cuda:0'), 'validation_loss_giou': tensor(0.0409, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3641, device='cuda:0'), 'train_loss_ce': tensor(0.2529, device='cuda:0'), 'train_loss_bbox': tensor(0.0059, device='cuda:0'), 'train_loss_giou': tensor(0.0409, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3907, device='cuda:0'), 'validation_loss_ce': tensor(0.2516, device='cuda:0'), 'validation_loss_bbox': tensor(0.0071, device='cuda:0'), 'validation_loss_giou': tensor(0.0519, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3907, device='cuda:0'), 'train_loss_ce': tensor(0.2516, device='cuda:0'), 'train_loss_bbox': tensor(0.0071, device='cuda:0'), 'train_loss_giou': tensor(0.0519, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3259, device='cuda:0'), 'validation_loss_ce': tensor(0.2508, device='cuda:0'), 'validation_loss_bbox': tensor(0.0047, device='cuda:0'), 'validation_loss_giou': tensor(0.0257, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3259, device='cuda:0'), 'train_loss_ce': tensor(0.2508, device='cuda:0'), 'train_loss_bbox': tensor(0.0047, device='cuda:0'), 'train_loss_giou': tensor(0.0257, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3595, device='cuda:0'), 'validation_loss_ce': tensor(0.2496, device='cuda:0'), 'validation_loss_bbox': tensor(0.0061, device='cuda:0'), 'validation_loss_giou': tensor(0.0398, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3595, device='cuda:0'), 'train_loss_ce': tensor(0.2496, device='cuda:0'), 'train_loss_bbox': tensor(0.0061, device='cuda:0'), 'train_loss_giou': tensor(0.0398, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3452, device='cuda:0'), 'validation_loss_ce': tensor(0.2482, device='cuda:0'), 'validation_loss_bbox': tensor(0.0053, device='cuda:0'), 'validation_loss_giou': tensor(0.0353, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3452, device='cuda:0'), 'train_loss_ce': tensor(0.2482, device='cuda:0'), 'train_loss_bbox': tensor(0.0053, device='cuda:0'), 'train_loss_giou': tensor(0.0353, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3760, device='cuda:0'), 'validation_loss_ce': tensor(0.2475, device='cuda:0'), 'validation_loss_bbox': tensor(0.0058, device='cuda:0'), 'validation_loss_giou': tensor(0.0498, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3760, device='cuda:0'), 'train_loss_ce': tensor(0.2475, device='cuda:0'), 'train_loss_bbox': tensor(0.0058, device='cuda:0'), 'train_loss_giou': tensor(0.0498, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3554, device='cuda:0'), 'validation_loss_ce': tensor(0.2468, device='cuda:0'), 'validation_loss_bbox': tensor(0.0054, device='cuda:0'), 'validation_loss_giou': tensor(0.0408, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3554, device='cuda:0'), 'train_loss_ce': tensor(0.2468, device='cuda:0'), 'train_loss_bbox': tensor(0.0054, device='cuda:0'), 'train_loss_giou': tensor(0.0408, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3424, device='cuda:0'), 'validation_loss_ce': tensor(0.2457, device='cuda:0'), 'validation_loss_bbox': tensor(0.0057, device='cuda:0'), 'validation_loss_giou': tensor(0.0341, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3424, device='cuda:0'), 'train_loss_ce': tensor(0.2457, device='cuda:0'), 'train_loss_bbox': tensor(0.0057, device='cuda:0'), 'train_loss_giou': tensor(0.0341, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3342, device='cuda:0'), 'validation_loss_ce': tensor(0.2450, device='cuda:0'), 'validation_loss_bbox': tensor(0.0038, device='cuda:0'), 'validation_loss_giou': tensor(0.0350, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3342, device='cuda:0'), 'train_loss_ce': tensor(0.2450, device='cuda:0'), 'train_loss_bbox': tensor(0.0038, device='cuda:0'), 'train_loss_giou': tensor(0.0350, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3353, device='cuda:0'), 'validation_loss_ce': tensor(0.2440, device='cuda:0'), 'validation_loss_bbox': tensor(0.0052, device='cuda:0'), 'validation_loss_giou': tensor(0.0326, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3353, device='cuda:0'), 'train_loss_ce': tensor(0.2440, device='cuda:0'), 'train_loss_bbox': tensor(0.0052, device='cuda:0'), 'train_loss_giou': tensor(0.0326, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3640, device='cuda:0'), 'validation_loss_ce': tensor(0.2434, device='cuda:0'), 'validation_loss_bbox': tensor(0.0059, device='cuda:0'), 'validation_loss_giou': tensor(0.0455, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3640, device='cuda:0'), 'train_loss_ce': tensor(0.2434, device='cuda:0'), 'train_loss_bbox': tensor(0.0059, device='cuda:0'), 'train_loss_giou': tensor(0.0455, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3303, device='cuda:0'), 'validation_loss_ce': tensor(0.2423, device='cuda:0'), 'validation_loss_bbox': tensor(0.0045, device='cuda:0'), 'validation_loss_giou': tensor(0.0328, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3303, device='cuda:0'), 'train_loss_ce': tensor(0.2423, device='cuda:0'), 'train_loss_bbox': tensor(0.0045, device='cuda:0'), 'train_loss_giou': tensor(0.0328, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3562, device='cuda:0'), 'validation_loss_ce': tensor(0.2409, device='cuda:0'), 'validation_loss_bbox': tensor(0.0049, device='cuda:0'), 'validation_loss_giou': tensor(0.0453, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3562, device='cuda:0'), 'train_loss_ce': tensor(0.2409, device='cuda:0'), 'train_loss_bbox': tensor(0.0049, device='cuda:0'), 'train_loss_giou': tensor(0.0453, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3637, device='cuda:0'), 'validation_loss_ce': tensor(0.2401, device='cuda:0'), 'validation_loss_bbox': tensor(0.0060, device='cuda:0'), 'validation_loss_giou': tensor(0.0467, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3637, device='cuda:0'), 'train_loss_ce': tensor(0.2401, device='cuda:0'), 'train_loss_bbox': tensor(0.0060, device='cuda:0'), 'train_loss_giou': tensor(0.0467, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3229, device='cuda:0'), 'validation_loss_ce': tensor(0.2389, device='cuda:0'), 'validation_loss_bbox': tensor(0.0052, device='cuda:0'), 'validation_loss_giou': tensor(0.0290, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3229, device='cuda:0'), 'train_loss_ce': tensor(0.2389, device='cuda:0'), 'train_loss_bbox': tensor(0.0052, device='cuda:0'), 'train_loss_giou': tensor(0.0290, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4256, device='cuda:0'), 'validation_loss_ce': tensor(0.2380, device='cuda:0'), 'validation_loss_bbox': tensor(0.0080, device='cuda:0'), 'validation_loss_giou': tensor(0.0739, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4256, device='cuda:0'), 'train_loss_ce': tensor(0.2380, device='cuda:0'), 'train_loss_bbox': tensor(0.0080, device='cuda:0'), 'train_loss_giou': tensor(0.0739, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4135, device='cuda:0'), 'validation_loss_ce': tensor(0.2368, device='cuda:0'), 'validation_loss_bbox': tensor(0.0084, device='cuda:0'), 'validation_loss_giou': tensor(0.0675, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4135, device='cuda:0'), 'train_loss_ce': tensor(0.2368, device='cuda:0'), 'train_loss_bbox': tensor(0.0084, device='cuda:0'), 'train_loss_giou': tensor(0.0675, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3890, device='cuda:0'), 'validation_loss_ce': tensor(0.2353, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0597, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3890, device='cuda:0'), 'train_loss_ce': tensor(0.2353, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0597, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4055, device='cuda:0'), 'validation_loss_ce': tensor(0.2341, device='cuda:0'), 'validation_loss_bbox': tensor(0.0075, device='cuda:0'), 'validation_loss_giou': tensor(0.0669, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4055, device='cuda:0'), 'train_loss_ce': tensor(0.2341, device='cuda:0'), 'train_loss_bbox': tensor(0.0075, device='cuda:0'), 'train_loss_giou': tensor(0.0669, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.4143, device='cuda:0'), 'validation_loss_ce': tensor(0.2333, device='cuda:0'), 'validation_loss_bbox': tensor(0.0091, device='cuda:0'), 'validation_loss_giou': tensor(0.0678, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.4143, device='cuda:0'), 'train_loss_ce': tensor(0.2333, device='cuda:0'), 'train_loss_bbox': tensor(0.0091, device='cuda:0'), 'train_loss_giou': tensor(0.0678, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3767, device='cuda:0'), 'validation_loss_ce': tensor(0.2325, device='cuda:0'), 'validation_loss_bbox': tensor(0.0077, device='cuda:0'), 'validation_loss_giou': tensor(0.0528, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3767, device='cuda:0'), 'train_loss_ce': tensor(0.2325, device='cuda:0'), 'train_loss_bbox': tensor(0.0077, device='cuda:0'), 'train_loss_giou': tensor(0.0528, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3804, device='cuda:0'), 'validation_loss_ce': tensor(0.2315, device='cuda:0'), 'validation_loss_bbox': tensor(0.0063, device='cuda:0'), 'validation_loss_giou': tensor(0.0588, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3804, device='cuda:0'), 'train_loss_ce': tensor(0.2315, device='cuda:0'), 'train_loss_bbox': tensor(0.0063, device='cuda:0'), 'train_loss_giou': tensor(0.0588, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3839, device='cuda:0'), 'validation_loss_ce': tensor(0.2306, device='cuda:0'), 'validation_loss_bbox': tensor(0.0061, device='cuda:0'), 'validation_loss_giou': tensor(0.0613, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3839, device='cuda:0'), 'train_loss_ce': tensor(0.2306, device='cuda:0'), 'train_loss_bbox': tensor(0.0061, device='cuda:0'), 'train_loss_giou': tensor(0.0613, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3984, device='cuda:0'), 'validation_loss_ce': tensor(0.2299, device='cuda:0'), 'validation_loss_bbox': tensor(0.0077, device='cuda:0'), 'validation_loss_giou': tensor(0.0650, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3984, device='cuda:0'), 'train_loss_ce': tensor(0.2299, device='cuda:0'), 'train_loss_bbox': tensor(0.0077, device='cuda:0'), 'train_loss_giou': tensor(0.0650, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3508, device='cuda:0'), 'validation_loss_ce': tensor(0.2290, device='cuda:0'), 'validation_loss_bbox': tensor(0.0074, device='cuda:0'), 'validation_loss_giou': tensor(0.0424, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3508, device='cuda:0'), 'train_loss_ce': tensor(0.2290, device='cuda:0'), 'train_loss_bbox': tensor(0.0074, device='cuda:0'), 'train_loss_giou': tensor(0.0424, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3513, device='cuda:0'), 'validation_loss_ce': tensor(0.2279, device='cuda:0'), 'validation_loss_bbox': tensor(0.0078, device='cuda:0'), 'validation_loss_giou': tensor(0.0421, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3513, device='cuda:0'), 'train_loss_ce': tensor(0.2279, device='cuda:0'), 'train_loss_bbox': tensor(0.0078, device='cuda:0'), 'train_loss_giou': tensor(0.0421, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3448, device='cuda:0'), 'validation_loss_ce': tensor(0.2266, device='cuda:0'), 'validation_loss_bbox': tensor(0.0069, device='cuda:0'), 'validation_loss_giou': tensor(0.0418, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3448, device='cuda:0'), 'train_loss_ce': tensor(0.2266, device='cuda:0'), 'train_loss_bbox': tensor(0.0069, device='cuda:0'), 'train_loss_giou': tensor(0.0418, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3470, device='cuda:0'), 'validation_loss_ce': tensor(0.2255, device='cuda:0'), 'validation_loss_bbox': tensor(0.0092, device='cuda:0'), 'validation_loss_giou': tensor(0.0377, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3470, device='cuda:0'), 'train_loss_ce': tensor(0.2255, device='cuda:0'), 'train_loss_bbox': tensor(0.0092, device='cuda:0'), 'train_loss_giou': tensor(0.0377, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3509, device='cuda:0'), 'validation_loss_ce': tensor(0.2244, device='cuda:0'), 'validation_loss_bbox': tensor(0.0091, device='cuda:0'), 'validation_loss_giou': tensor(0.0405, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3509, device='cuda:0'), 'train_loss_ce': tensor(0.2244, device='cuda:0'), 'train_loss_bbox': tensor(0.0091, device='cuda:0'), 'train_loss_giou': tensor(0.0405, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3716, device='cuda:0'), 'validation_loss_ce': tensor(0.2235, device='cuda:0'), 'validation_loss_bbox': tensor(0.0095, device='cuda:0'), 'validation_loss_giou': tensor(0.0503, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3716, device='cuda:0'), 'train_loss_ce': tensor(0.2235, device='cuda:0'), 'train_loss_bbox': tensor(0.0095, device='cuda:0'), 'train_loss_giou': tensor(0.0503, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3575, device='cuda:0'), 'validation_loss_ce': tensor(0.2222, device='cuda:0'), 'validation_loss_bbox': tensor(0.0099, device='cuda:0'), 'validation_loss_giou': tensor(0.0429, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3575, device='cuda:0'), 'train_loss_ce': tensor(0.2222, device='cuda:0'), 'train_loss_bbox': tensor(0.0099, device='cuda:0'), 'train_loss_giou': tensor(0.0429, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3330, device='cuda:0'), 'validation_loss_ce': tensor(0.2211, device='cuda:0'), 'validation_loss_bbox': tensor(0.0078, device='cuda:0'), 'validation_loss_giou': tensor(0.0364, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3330, device='cuda:0'), 'train_loss_ce': tensor(0.2211, device='cuda:0'), 'train_loss_bbox': tensor(0.0078, device='cuda:0'), 'train_loss_giou': tensor(0.0364, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3988, device='cuda:0'), 'validation_loss_ce': tensor(0.2201, device='cuda:0'), 'validation_loss_bbox': tensor(0.0126, device='cuda:0'), 'validation_loss_giou': tensor(0.0579, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+{'training_loss': tensor(0.3988, device='cuda:0'), 'train_loss_ce': tensor(0.2201, device='cuda:0'), 'train_loss_bbox': tensor(0.0126, device='cuda:0'), 'train_loss_giou': tensor(0.0579, device='cuda:0'), 'train_cardinality_error': tensor(1., device='cuda:0'), 'validation_loss': tensor(0.3686, device='cuda:0'), 'validation_loss_ce': tensor(0.2194, device='cuda:0'), 'validation_loss_bbox': tensor(0.0093, device='cuda:0'), 'validation_loss_giou': tensor(0.0514, device='cuda:0'), 'validation_cardinality_error': tensor(1., device='cuda:0')}
+```