racoon

This model is a fine-tuned version of hustvl/yolos-tiny on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3297
• Map Raccoon: 0.2841
• Mar 100 Raccoon: 1.0

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 8
• 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: cosine
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Map Raccoon	Mar 100 Raccoon
No log	1.0	50	1.6101	0.2337	0.9979
No log	2.0	100	1.5579	0.2460	0.9976
No log	3.0	150	1.4528	0.2138	0.9977
No log	4.0	200	1.0310	0.2317	0.9996
No log	5.0	250	0.8957	0.2477	0.9996
No log	6.0	300	0.7868	0.2655	0.9998
No log	7.0	350	0.7721	0.2117	0.9998
No log	8.0	400	0.7410	0.3318	0.9998
No log	9.0	450	0.6789	0.2537	0.9998
1.0217	10.0	500	0.6666	0.2094	0.9998
1.0217	11.0	550	0.5995	0.2813	0.9998
1.0217	12.0	600	0.5440	0.2875	0.9998
1.0217	13.0	650	0.5423	0.2558	0.9998
1.0217	14.0	700	0.5304	0.2311	1.0
1.0217	15.0	750	0.4702	0.2450	1.0
1.0217	16.0	800	0.4621	0.2278	0.9998
1.0217	17.0	850	0.4398	0.227	1.0
1.0217	18.0	900	0.4342	0.2384	1.0
1.0217	19.0	950	0.4023	0.2525	1.0
0.5565	20.0	1000	0.3909	0.2728	1.0
0.5565	21.0	1050	0.4083	0.2868	1.0
0.5565	22.0	1100	0.3783	0.2726	1.0
0.5565	23.0	1150	0.3564	0.2811	1.0
0.5565	24.0	1200	0.3467	0.2764	1.0
0.5565	25.0	1250	0.3411	0.2774	1.0
0.5565	26.0	1300	0.3346	0.2799	1.0
0.5565	27.0	1350	0.3315	0.2807	1.0
0.5565	28.0	1400	0.3290	0.2829	1.0
0.5565	29.0	1450	0.3299	0.2842	1.0
0.4089	30.0	1500	0.3297	0.2841	1.0

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1