chickens-300-epoch-200-images

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

• Loss: 0.5365

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: 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
1.9097	1.0	99	1.7611
1.6037	2.0	198	1.5045
1.5109	3.0	297	1.3585
1.3272	4.0	396	1.2303
1.3035	5.0	495	1.2142
1.1962	6.0	594	1.1744
1.1761	7.0	693	1.0835
1.1091	8.0	792	1.0690
1.1074	9.0	891	1.0030
0.9913	10.0	990	0.9990
0.9685	11.0	1089	0.9787
0.9439	12.0	1188	0.9659
0.9102	13.0	1287	0.9237
0.9849	14.0	1386	0.8619
0.8692	15.0	1485	0.8463
0.833	16.0	1584	0.8279
0.8148	17.0	1683	0.7647
0.7309	18.0	1782	0.8500
0.7024	19.0	1881	0.7294
0.73	20.0	1980	0.7389
0.7115	21.0	2079	0.7617
0.6883	22.0	2178	0.8105
0.7151	23.0	2277	0.6843
0.6598	24.0	2376	0.7199
0.762	25.0	2475	0.6930
0.6331	26.0	2574	0.6905
0.6141	27.0	2673	0.6327
0.6014	28.0	2772	0.6516
0.5935	29.0	2871	0.6052
0.557	30.0	2970	0.5846
0.6292	31.0	3069	0.5857
0.5536	32.0	3168	0.5905
0.5309	33.0	3267	0.6030
0.5065	34.0	3366	0.5743
0.5429	35.0	3465	0.5699
0.4981	36.0	3564	0.5755
0.5184	37.0	3663	0.5689
0.5338	38.0	3762	0.5646
0.5218	39.0	3861	0.5688
0.5102	40.0	3960	0.5738
0.5052	41.0	4059	0.5599
0.4897	42.0	4158	0.5454
0.4916	43.0	4257	0.5393
0.4696	44.0	4356	0.5470
0.494	45.0	4455	0.5516
0.465	46.0	4554	0.5459
0.4733	47.0	4653	0.5483
0.4929	48.0	4752	0.5316
0.4822	49.0	4851	0.5306
0.4594	50.0	4950	0.5351
0.4598	51.0	5049	0.5348
0.4511	52.0	5148	0.5432
0.4561	53.0	5247	0.5467
0.4653	54.0	5346	0.5444
0.4475	55.0	5445	0.5423
0.4734	56.0	5544	0.5368
0.4815	57.0	5643	0.5363
0.4365	58.0	5742	0.5366
0.4596	59.0	5841	0.5365
0.4757	60.0	5940	0.5365

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 2.19.2
• Tokenizers 0.20.0