fisheye8k_microsoft_conditional-detr-resnet-50

This model is a fine-tuned version of microsoft/conditional-detr-resnet-50 on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 1.4466

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 8
• seed: 0
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 36
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.0211	1.0	5288	1.5012
0.9117	2.0	10576	1.4713
0.8595	3.0	15864	1.4364
0.7922	4.0	21152	1.5227
0.7764	5.0	26440	1.6631
0.7419	6.0	31728	1.4320
0.7132	7.0	37016	1.4661
0.6991	8.0	42304	1.4318
0.6585	9.0	47592	1.4069
0.6527	10.0	52880	1.4213
0.6191	11.0	58168	1.4144
0.6248	12.0	63456	1.3887
0.6085	13.0	68744	1.4053
0.582	14.0	74032	1.4418
0.5592	15.0	79320	1.5815
0.552	16.0	84608	1.4832
0.5233	17.0	89896	1.4466

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0