detr-amzss3-v2

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: 0.3494

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.54	1000	0.4810
0.5325	1.08	2000	0.4812
0.5325	1.62	3000	0.4739
0.5322	2.16	4000	0.4759
0.5322	2.7	5000	0.4818
0.5259	3.24	6000	0.4522
0.5259	3.78	7000	0.4632
0.5167	4.32	8000	0.4628
0.5167	4.86	9000	0.4345
0.5076	5.4	10000	0.4563
0.5076	5.94	11000	0.4326
0.494	6.48	12000	0.4424
0.4906	7.02	13000	0.4272
0.4906	7.56	14000	0.4164
0.4801	8.1	15000	0.4213
0.4801	8.64	16000	0.4320
0.4699	9.18	17000	0.4100
0.4699	9.72	18000	0.4127
0.4613	10.26	19000	0.4035
0.4613	10.8	20000	0.4039
0.4556	11.34	21000	0.4149
0.4556	11.88	22000	0.4092
0.4475	12.42	23000	0.3965
0.4475	12.96	24000	0.3973
0.4389	13.5	25000	0.4013
0.4349	14.04	26000	0.3797
0.4349	14.58	27000	0.3728
0.4288	15.12	28000	0.3834
0.4288	15.66	29000	0.3885
0.4222	16.2	30000	0.3820
0.4222	16.74	31000	0.3755
0.4152	17.28	32000	0.3693
0.4152	17.82	33000	0.3679
0.4122	18.36	34000	0.3605
0.4122	18.9	35000	0.3625
0.4077	19.44	36000	0.3631
0.4077	19.98	37000	0.3607
0.4	20.52	38000	0.3615
0.3972	21.06	39000	0.3561
0.3972	21.6	40000	0.3594
0.3953	22.14	41000	0.3554
0.3953	22.68	42000	0.3515
0.3903	23.22	43000	0.3539
0.3903	23.76	44000	0.3500
0.3878	24.3	45000	0.3489
0.3878	24.84	46000	0.3494

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3