MT-proud-rain-95

This model is a fine-tuned version of toobiza/MT-ancient-spaceship-83 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1562
• Loss Ce: 0.0000
• Loss Bbox: 0.0213
• Cardinality Error: 1.0
• Giou: 97.5230

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Loss Ce	Loss Bbox	Cardinality Error	Giou
0.1665	0.48	200	0.2101	0.0000	0.0293	1.0	96.8141
0.1967	0.97	400	0.1844	0.0000	0.0255	1.0	97.1659
0.1624	1.45	600	0.1833	0.0000	0.0253	1.0	97.1706
0.1594	1.93	800	0.1720	0.0000	0.0237	1.0	97.3363
0.1598	2.42	1000	0.1729	0.0000	0.0238	1.0	97.3105
0.1941	2.9	1200	0.1494	0.0000	0.0203	1.0	97.6099
0.1223	3.38	1400	0.1525	0.0000	0.0209	1.0	97.6036
0.1514	3.86	1600	0.1512	0.0000	0.0207	1.0	97.6045
0.1585	4.35	1800	0.1569	0.0000	0.0215	1.0	97.5391
0.128	4.83	2000	0.1535	0.0000	0.0210	1.0	97.5658
0.1089	5.31	2200	0.1594	0.0000	0.0220	1.0	97.5180
0.1624	5.8	2400	0.1650	0.0000	0.0228	1.0	97.4441
0.1074	6.28	2600	0.1648	0.0000	0.0227	1.0	97.4209
0.1693	6.76	2800	0.1554	0.0000	0.0212	1.0	97.5341
0.1075	7.25	3000	0.1595	0.0000	0.0218	1.0	97.4777
0.1271	7.73	3200	0.1570	0.0000	0.0215	1.0	97.5156
0.1293	8.21	3400	0.1549	0.0000	0.0211	1.0	97.5331
0.1143	8.7	3600	0.1564	0.0000	0.0214	1.0	97.5335
0.0966	9.18	3800	0.1555	0.0000	0.0213	1.0	97.5400
0.104	9.66	4000	0.1562	0.0000	0.0213	1.0	97.5230

Framework versions

• Transformers 4.33.2
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.13.3