metadata

license: apache-2.0
base_model: WKLI22/detr-resnet-50_finetuned_cppe5
tags:
  - generated_from_trainer
model-index:
  - name: detr-resnet-50_finetuned_cppe5
    results: []

detr-resnet-50_finetuned_cppe5

This model is a fine-tuned version of WKLI22/detr-resnet-50_finetuned_cppe5 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.4986

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: 17
eval_batch_size: 17
seed: 42
gradient_accumulation_steps: 6
total_train_batch_size: 102
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4724	0.07	2	0.5355
0.4907	0.14	4	0.5277
0.498	0.21	6	0.5214
0.534	0.28	8	0.5274
0.5305	0.36	10	0.5294
0.5096	0.43	12	0.5273
0.5149	0.5	14	0.5158
0.5099	0.57	16	0.5163
0.5308	0.64	18	0.5217
0.5164	0.71	20	0.5076
0.4932	0.78	22	0.5050
0.5151	0.85	24	0.5102
0.4982	0.92	26	0.5044
0.526	0.99	28	0.5096
0.5034	1.07	30	0.4980
0.5155	1.14	32	0.5067
0.513	1.21	34	0.5011
0.5019	1.28	36	0.5066
0.4704	1.35	38	0.5094
0.5404	1.42	40	0.5126
0.5263	1.49	42	0.5062
0.4729	1.56	44	0.5223
0.5032	1.63	46	0.5073
0.476	1.7	48	0.5111
0.4823	1.78	50	0.5094
0.5223	1.85	52	0.5042
0.4855	1.92	54	0.4962
0.5038	1.99	56	0.5006
0.5196	2.06	58	0.5022
0.4847	2.13	60	0.4943
0.4697	2.2	62	0.5007
0.4893	2.27	64	0.5041
0.4939	2.34	66	0.4910
0.5093	2.41	68	0.4974
0.4884	2.49	70	0.4962
0.5087	2.56	72	0.5081
0.4889	2.63	74	0.4974
0.486	2.7	76	0.5003
0.4747	2.77	78	0.4917
0.5028	2.84	80	0.4849
0.5366	2.91	82	0.4931
0.5083	2.98	84	0.4986

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2