detr-resnet-50-dc5-grasshopper-finetuned-maxsteps-1000

This model is a fine-tuned version of facebook/detr-resnet-50-dc5 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.0027
• Map: 0.0011
• Map 50: 0.0042
• Map 75: 0.0001
• Map Small: 0.0011
• Map Medium: -1.0
• Map Large: -1.0
• Mar 1: 0.0005
• Mar 10: 0.0038
• Mar 100: 0.0397
• Mar Small: 0.0397
• Mar Medium: -1.0
• Mar Large: -1.0
• Map Recilia dorsalis: 0.001
• Mar 100 Recilia dorsalis: 0.029
• Map Nephotettix malayanus: 0.0033
• Mar 100 Nephotettix malayanus: 0.1299
• Map Sogatella furcifera: 0.0
• Mar 100 Sogatella furcifera: 0.0
• Map Nilaparvata lugens: 0.0
• Mar 100 Nilaparvata lugens: 0.0

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Recilia dorsalis	Mar 100 Recilia dorsalis	Map Nephotettix malayanus	Mar 100 Nephotettix malayanus	Map Sogatella furcifera	Mar 100 Sogatella furcifera	Map Nilaparvata lugens	Mar 100 Nilaparvata lugens
4.266	2.3810	50	3.2196	0.0007	0.0035	0.0001	0.0007	-1.0	-1.0	0.0	0.0029	0.0298	0.0298	-1.0	-1.0	0.0007	0.0234	0.0022	0.0959	0.0	0.0	0.0	0.0
3.8932	4.7619	100	3.0027	0.0011	0.0042	0.0001	0.0011	-1.0	-1.0	0.0005	0.0038	0.0397	0.0397	-1.0	-1.0	0.001	0.029	0.0033	0.1299	0.0	0.0	0.0	0.0

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1
• Datasets 3.2.0
• Tokenizers 0.20.3