practica_2_model2

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: 0.3766

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	17	0.9447
No log	2.0	34	0.9111
1.0504	3.0	51	0.7995
1.0504	4.0	68	0.7310
1.0504	5.0	85	0.6850
0.7808	6.0	102	0.6665
0.7808	7.0	119	0.6733
0.7808	8.0	136	0.5836
0.6128	9.0	153	0.5371
0.6128	10.0	170	0.5584
0.6128	11.0	187	0.5198
0.504	12.0	204	0.4960
0.504	13.0	221	0.4524
0.504	14.0	238	0.5036
0.4324	15.0	255	0.4910
0.4324	16.0	272	0.4838
0.4324	17.0	289	0.4706
0.3837	18.0	306	0.4151
0.3837	19.0	323	0.4629
0.3837	20.0	340	0.4957
0.3792	21.0	357	0.3997
0.3792	22.0	374	0.5092
0.3792	23.0	391	0.4898
0.3622	24.0	408	0.4457
0.3622	25.0	425	0.4580
0.3622	26.0	442	0.4016
0.3735	27.0	459	0.4088
0.3735	28.0	476	0.4354
0.3735	29.0	493	0.4236
0.3106	30.0	510	0.4381
0.3106	31.0	527	0.3804
0.3106	32.0	544	0.3955
0.3048	33.0	561	0.4086
0.3048	34.0	578	0.4342
0.3048	35.0	595	0.3663
0.2901	36.0	612	0.4711
0.2901	37.0	629	0.4031
0.2901	38.0	646	0.3623
0.2798	39.0	663	0.4112
0.2798	40.0	680	0.3581
0.2798	41.0	697	0.3723
0.2669	42.0	714	0.4367
0.2669	43.0	731	0.3548
0.2669	44.0	748	0.3789
0.2486	45.0	765	0.3873
0.2486	46.0	782	0.3687
0.2486	47.0	799	0.3990
0.2401	48.0	816	0.3698
0.2401	49.0	833	0.3697
0.2432	50.0	850	0.4309
0.2432	51.0	867	0.3439
0.2432	52.0	884	0.3556
0.2319	53.0	901	0.3514
0.2319	54.0	918	0.3402
0.2319	55.0	935	0.3706
0.2179	56.0	952	0.3511
0.2179	57.0	969	0.3479
0.2179	58.0	986	0.4039
0.2192	59.0	1003	0.3684
0.2192	60.0	1020	0.3531
0.2192	61.0	1037	0.3475
0.2087	62.0	1054	0.3374
0.2087	63.0	1071	0.3321
0.2087	64.0	1088	0.3440
0.2001	65.0	1105	0.3741
0.2001	66.0	1122	0.3493
0.2001	67.0	1139	0.4159
0.1992	68.0	1156	0.3376
0.1992	69.0	1173	0.3495
0.1992	70.0	1190	0.3530
0.1899	71.0	1207	0.3254
0.1899	72.0	1224	0.3796
0.1899	73.0	1241	0.3437
0.1861	74.0	1258	0.3489
0.1861	75.0	1275	0.3398
0.1861	76.0	1292	0.3929
0.1896	77.0	1309	0.3382
0.1896	78.0	1326	0.3609
0.1896	79.0	1343	0.3501
0.1859	80.0	1360	0.3392
0.1859	81.0	1377	0.3448
0.1859	82.0	1394	0.3621
0.1808	83.0	1411	0.3525
0.1808	84.0	1428	0.3553
0.1808	85.0	1445	0.3645
0.1843	86.0	1462	0.3306
0.1843	87.0	1479	0.3627
0.1843	88.0	1496	0.3497
0.1743	89.0	1513	0.3892
0.1743	90.0	1530	0.3410
0.1743	91.0	1547	0.3568
0.1689	92.0	1564	0.3409
0.1689	93.0	1581	0.3505
0.1689	94.0	1598	0.3536
0.1675	95.0	1615	0.3792
0.1675	96.0	1632	0.3321
0.1675	97.0	1649	0.3489
0.1734	98.0	1666	0.3374
0.1734	99.0	1683	0.3417
0.1689	100.0	1700	0.3766

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0