MT-ancient-spaceship-83

Se utilizaron los siguientes hiperparámetros durante el entrenamiento:

- learning_rate: 1e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam con betas=(0.9,0.999) y epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3

Resultados de Entrenamiento:

- Epoch 0.35, Step 50, Validation Loss: 0.1849, Loss Ce: 0.0000, Loss Bbox: 0.0252, Cardinality Error: 1.0, Giou: 97.2490
- Epoch 0.69, Step 100, Validation Loss: 0.1683, Loss Ce: 0.0000, Loss Bbox: 0.0231, Cardinality Error: 1.0, Giou: 97.4814
- Epoch 1.04, Step 150, Validation Loss: 0.1701, Loss Ce: 0.0000, Loss Bbox: 0.0235, Cardinality Error: 1.0, Giou: 97.4533
- Epoch 1.39, Step 200, Validation Loss: 0.1864, Loss Ce: 0.0000, Loss Bbox: 0.0259, Cardinality Error: 1.0, Giou: 97.2923
- Epoch 1.74, Step 250, Validation Loss: 0.1786, Loss Ce: 0.0000, Loss Bbox: 0.0246, Cardinality Error: 1.0, Giou: 97.4046
- Epoch 2.08, Step 300, Validation Loss: 0.1673, Loss Ce: 0.0000, Loss Bbox: 0.0230, Cardinality Error: 1.0, Giou: 97.5302
- Epoch 2.43, Step 350, Validation Loss: 0.1710, Loss Ce: 0.0000, Loss Bbox: 0.0235, Cardinality Error: 1.0, Giou: 97.4826
- Epoch 2.78, Step 400, Validation Loss: 0.1689, Loss Ce: 0.0000, Loss Bbox: 0.0233, Cardinality Error: 1.0, Giou: 97.5276