chickens-repro-4

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

• Loss: 0.2536

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: 2
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss
1.2251	1.0	227	1.1143
1.1432	2.0	454	1.3992
1.0496	3.0	681	0.9853
0.9896	4.0	908	0.7757
0.8727	5.0	1135	0.7749
0.7478	6.0	1362	0.7531
0.694	7.0	1589	0.5933
0.7923	8.0	1816	0.6223
0.8413	9.0	2043	0.6116
0.8007	10.0	2270	0.8077
0.7769	11.0	2497	0.7307
0.7564	12.0	2724	0.6538
0.7178	13.0	2951	0.5011
0.724	14.0	3178	0.7358
0.7586	15.0	3405	0.6229
0.7284	16.0	3632	0.6627
0.6092	17.0	3859	0.5987
0.6531	18.0	4086	0.6447
0.5879	19.0	4313	0.4965
0.62	20.0	4540	0.4803
0.5851	21.0	4767	0.5303
0.5981	22.0	4994	0.4260
0.5593	23.0	5221	0.4870
0.6139	24.0	5448	0.4931
0.5381	25.0	5675	0.4639
0.5569	26.0	5902	0.4438
0.5432	27.0	6129	0.4747
0.5557	28.0	6356	0.4214
0.4997	29.0	6583	0.4308
0.5008	30.0	6810	0.3995
0.4966	31.0	7037	0.3612
0.4636	32.0	7264	0.4262
0.4566	33.0	7491	0.3896
0.4301	34.0	7718	0.3916
0.4501	35.0	7945	0.3290
0.4454	36.0	8172	0.3854
0.4313	37.0	8399	0.3072
0.4004	38.0	8626	0.3387
0.3943	39.0	8853	0.3349
0.397	40.0	9080	0.2835
0.3962	41.0	9307	0.2743
0.3924	42.0	9534	0.3099
0.373	43.0	9761	0.2811
0.3567	44.0	9988	0.2673
0.3675	45.0	10215	0.2863
0.3687	46.0	10442	0.2822
0.3495	47.0	10669	0.2650
0.3413	48.0	10896	0.2669
0.3495	49.0	11123	0.2664
0.3208	50.0	11350	0.2788
0.3348	51.0	11577	0.2763
0.34	52.0	11804	0.2615
0.3595	53.0	12031	0.2553
0.3319	54.0	12258	0.2588
0.342	55.0	12485	0.2597
0.3329	56.0	12712	0.2542
0.3357	57.0	12939	0.2543
0.3364	58.0	13166	0.2504
0.3278	59.0	13393	0.2535
0.3351	60.0	13620	0.2536

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1