chickens-repro

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: 0.1629

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
1.0965	1.0	227	1.0412
0.6989	2.0	454	0.5108
0.6796	3.0	681	0.5768
0.5676	4.0	908	0.5100
0.5979	5.0	1135	0.4557
0.5055	6.0	1362	0.4868
0.4916	7.0	1589	0.3862
0.5587	8.0	1816	0.3914
0.4772	9.0	2043	0.3713
0.5193	10.0	2270	0.3641
0.5099	11.0	2497	0.4205
0.581	12.0	2724	0.4537
0.52	13.0	2951	0.4301
0.4794	14.0	3178	0.4041
0.5919	15.0	3405	0.6219
0.5318	16.0	3632	0.4846
0.5512	17.0	3859	0.5291
0.4638	18.0	4086	0.3529
0.4342	19.0	4313	0.3357
0.3997	20.0	4540	0.3045
0.4364	21.0	4767	0.4230
0.4158	22.0	4994	0.4313
0.408	23.0	5221	0.3280
0.4173	24.0	5448	0.3922
0.3976	25.0	5675	0.3159
0.4172	26.0	5902	0.3246
0.3991	27.0	6129	0.3080
0.4005	28.0	6356	0.3559
0.3632	29.0	6583	0.3132
0.3844	30.0	6810	0.3978
0.3685	31.0	7037	0.3134
0.3339	32.0	7264	0.2840
0.3398	33.0	7491	0.2782
0.3233	34.0	7718	0.2905
0.322	35.0	7945	0.2249
0.3196	36.0	8172	0.2325
0.319	37.0	8399	0.2156
0.3092	38.0	8626	0.2010
0.2953	39.0	8853	0.2158
0.3245	40.0	9080	0.2274
0.2881	41.0	9307	0.2449
0.3029	42.0	9534	0.2069
0.3104	43.0	9761	0.2417
0.2754	44.0	9988	0.1988
0.2746	45.0	10215	0.1900
0.2695	46.0	10442	0.1792
0.2639	47.0	10669	0.1795
0.2484	48.0	10896	0.1971
0.2585	49.0	11123	0.1694
0.2489	50.0	11350	0.1886
0.2525	51.0	11577	0.1756
0.2517	52.0	11804	0.1704
0.2375	53.0	12031	0.1697
0.2496	54.0	12258	0.1731
0.2476	55.0	12485	0.1665
0.2381	56.0	12712	0.1596
0.2249	57.0	12939	0.1669
0.2401	58.0	13166	0.1618
0.2323	59.0	13393	0.1623
0.2482	60.0	13620	0.1629

Framework versions

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