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: 0.3677

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: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 96
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3835	0.17	5	0.4158
0.4103	0.34	10	0.4217
0.4363	0.5	15	0.3950
0.3906	0.67	20	0.4011
0.3773	0.84	25	0.4191
0.4286	1.01	30	0.4231
0.4122	1.17	35	0.3977
0.3845	1.34	40	0.3812
0.4228	1.51	45	0.3750
0.4139	1.68	50	0.4094
0.433	1.84	55	0.3887
0.3752	2.01	60	0.4064
0.3835	2.18	65	0.3992
0.3756	2.35	70	0.3898
0.3918	2.52	75	0.3946
0.3715	2.68	80	0.3972
0.379	2.85	85	0.3951
0.4104	3.02	90	0.3926
0.3769	3.19	95	0.3888
0.3807	3.35	100	0.3956
0.3737	3.52	105	0.3989
0.3634	3.69	110	0.3978
0.3922	3.86	115	0.4041
0.3768	4.03	120	0.3846
0.4015	4.19	125	0.3973
0.3655	4.36	130	0.3879
0.3934	4.53	135	0.3775
0.3586	4.7	140	0.3851
0.3628	4.86	145	0.3785
0.3859	5.03	150	0.3922
0.3938	5.2	155	0.3982
0.3766	5.37	160	0.3923
0.3622	5.53	165	0.3960
0.3632	5.7	170	0.4003
0.3569	5.87	175	0.3924
0.3461	6.04	180	0.3972
0.3745	6.21	185	0.3938
0.358	6.37	190	0.3952
0.3671	6.54	195	0.3964
0.3605	6.71	200	0.4059
0.3764	6.88	205	0.4024
0.3477	7.04	210	0.3999
0.3493	7.21	215	0.3952
0.4103	7.38	220	0.4001
0.3693	7.55	225	0.3965
0.3892	7.71	230	0.4026
0.3719	7.88	235	0.3945
0.3793	8.05	240	0.3907
0.3563	8.22	245	0.3937
0.3361	8.39	250	0.3896
0.3563	8.55	255	0.3968
0.3482	8.72	260	0.3953
0.3511	8.89	265	0.3808
0.3654	9.06	270	0.3754
0.3315	9.22	275	0.3946
0.3416	9.39	280	0.3812
0.3527	9.56	285	0.3842
0.3568	9.73	290	0.3888
0.3502	9.9	295	0.3798
0.3676	10.06	300	0.3886
0.3288	10.23	305	0.3976
0.3465	10.4	310	0.3919
0.329	10.57	315	0.3894
0.3413	10.73	320	0.3877
0.3315	10.9	325	0.3942
0.3336	11.07	330	0.3885
0.3203	11.24	335	0.3879
0.3331	11.4	340	0.3878
0.3562	11.57	345	0.3810
0.3262	11.74	350	0.3787
0.3185	11.91	355	0.3902
0.353	12.08	360	0.3885
0.3388	12.24	365	0.3842
0.3158	12.41	370	0.3809
0.3596	12.58	375	0.3876
0.342	12.75	380	0.3796
0.3295	12.91	385	0.3781
0.3154	13.08	390	0.3837
0.3233	13.25	395	0.3765
0.3314	13.42	400	0.3813
0.3291	13.58	405	0.3813
0.3334	13.75	410	0.3625
0.3392	13.92	415	0.3600
0.3053	14.09	420	0.3637
0.3501	14.26	425	0.3567
0.3426	14.42	430	0.3775
0.3581	14.59	435	0.3658
0.3177	14.76	440	0.3840
0.3241	14.93	445	0.3802
0.3206	15.09	450	0.3903
0.3341	15.26	455	0.3749
0.3388	15.43	460	0.3735
0.3353	15.6	465	0.3718
0.341	15.77	470	0.3808
0.3471	15.93	475	0.3911
0.332	16.1	480	0.3900
0.3132	16.27	485	0.3883
0.3307	16.44	490	0.3899
0.3014	16.6	495	0.3757
0.3271	16.77	500	0.3730
0.3324	16.94	505	0.3678
0.3203	17.11	510	0.3603
0.3192	17.27	515	0.3727
0.3281	17.44	520	0.4148
0.3302	17.61	525	0.3644
0.3109	17.78	530	0.3890
0.3606	17.95	535	0.3812
0.306	18.11	540	0.3622
0.3249	18.28	545	0.3559
0.3179	18.45	550	0.3595
0.3018	18.62	555	0.3686
0.2852	18.78	560	0.3501
0.3306	18.95	565	0.3684
0.294	19.12	570	0.3661
0.3103	19.29	575	0.3662
0.3123	19.45	580	0.3397
0.292	19.62	585	0.3691
0.3186	19.79	590	0.3758
0.3062	19.96	595	0.3509
0.292	20.13	600	0.3803
0.3363	20.29	605	0.3527
0.3067	20.46	610	0.3611
0.317	20.63	615	0.3676
0.3043	20.8	620	0.3676
0.3642	20.96	625	0.3785
0.3279	21.13	630	0.3957
0.3208	21.3	635	0.3788
0.3425	21.47	640	0.3791
0.2832	21.64	645	0.3816
0.3149	21.8	650	0.3693
0.312	21.97	655	0.3709
0.3276	22.14	660	0.3710
0.2862	22.31	665	0.3742
0.3234	22.47	670	0.3776
0.2984	22.64	675	0.3933
0.3204	22.81	680	0.3827
0.3107	22.98	685	0.3850
0.2908	23.14	690	0.3780
0.3171	23.31	695	0.3785
0.3248	23.48	700	0.3671
0.3164	23.65	705	0.3646
0.3249	23.82	710	0.3652
0.3001	23.98	715	0.3621
0.3039	24.15	720	0.3743
0.2927	24.32	725	0.3740
0.2946	24.49	730	0.4031
0.3636	24.65	735	0.3715
0.2915	24.82	740	0.3737
0.3202	24.99	745	0.3805
0.2904	25.16	750	0.3841
0.3031	25.32	755	0.3907
0.2904	25.49	760	0.3709
0.3098	25.66	765	0.3806
0.297	25.83	770	0.3782
0.3062	26.0	775	0.3918
0.2856	26.16	780	0.3835
0.3209	26.33	785	0.4070
0.2959	26.5	790	0.3846
0.2812	26.67	795	0.3867
0.3027	26.83	800	0.3684
0.2942	27.0	805	0.3708
0.2754	27.17	810	0.3468
0.31	27.34	815	0.3463
0.3064	27.51	820	0.3296
0.2953	27.67	825	0.3416
0.3252	27.84	830	0.3556
0.2949	28.01	835	0.3662
0.3144	28.18	840	0.3682
0.3045	28.34	845	0.3501
0.3298	28.51	850	0.3645
0.2961	28.68	855	0.3744
0.317	28.85	860	0.3807
0.3213	29.01	865	0.3620
0.297	29.18	870	0.3657
0.3132	29.35	875	0.3635
0.3002	29.52	880	0.3697
0.2904	29.69	885	0.3600
0.2853	29.85	890	0.3670
0.3265	30.02	895	0.3693
0.2964	30.19	900	0.3586
0.2842	30.36	905	0.3550
0.3038	30.52	910	0.3686
0.2875	30.69	915	0.3705
0.2871	30.86	920	0.3586
0.311	31.03	925	0.3527
0.2813	31.19	930	0.3532
0.3227	31.36	935	0.3600
0.3063	31.53	940	0.3743
0.2929	31.7	945	0.3557
0.2946	31.87	950	0.3723
0.3161	32.03	955	0.3578
0.3067	32.2	960	0.3735
0.3052	32.37	965	0.3584
0.2924	32.54	970	0.3587
0.3011	32.7	975	0.3923
0.2727	32.87	980	0.3645
0.3052	33.04	985	0.3843
0.2859	33.21	990	0.3675
0.3002	33.38	995	0.3703
0.2814	33.54	1000	0.3718
0.2935	33.71	1005	0.3766
0.3045	33.88	1010	0.3943
0.3116	34.05	1015	0.3996
0.288	34.21	1020	0.3897
0.2894	34.38	1025	0.3844
0.2773	34.55	1030	0.3639
0.2785	34.72	1035	0.3808
0.2973	34.88	1040	0.3832
0.2746	35.05	1045	0.3745
0.27	35.22	1050	0.3754
0.2895	35.39	1055	0.3727
0.3008	35.56	1060	0.3589
0.2808	35.72	1065	0.3617
0.2991	35.89	1070	0.3672
0.2941	36.06	1075	0.3735
0.2722	36.23	1080	0.3621
0.3002	36.39	1085	0.3759
0.2851	36.56	1090	0.3666
0.2832	36.73	1095	0.3617
0.2979	36.9	1100	0.3736
0.3136	37.06	1105	0.3755
0.2867	37.23	1110	0.3634
0.2891	37.4	1115	0.3671
0.2947	37.57	1120	0.3772
0.2812	37.74	1125	0.3647
0.2958	37.9	1130	0.3618
0.2829	38.07	1135	0.3812
0.3064	38.24	1140	0.3684
0.2851	38.41	1145	0.3865
0.2888	38.57	1150	0.3849
0.2772	38.74	1155	0.3930
0.287	38.91	1160	0.3837
0.3077	39.08	1165	0.3699
0.2701	39.25	1170	0.3682
0.2651	39.41	1175	0.3711
0.2935	39.58	1180	0.3974
0.2804	39.75	1185	0.3863
0.28	39.92	1190	0.4039
0.2836	40.08	1195	0.3900
0.2942	40.25	1200	0.3921
0.2881	40.42	1205	0.3855
0.2975	40.59	1210	0.3711
0.2975	40.75	1215	0.3805
0.2637	40.92	1220	0.4088
0.2761	41.09	1225	0.4052
0.2966	41.26	1230	0.3882
0.2699	41.43	1235	0.3987
0.286	41.59	1240	0.3958
0.284	41.76	1245	0.4042
0.2769	41.93	1250	0.3995
0.2868	42.1	1255	0.3761
0.2869	42.26	1260	0.3920
0.2953	42.43	1265	0.3869
0.2647	42.6	1270	0.4168
0.2737	42.77	1275	0.3879
0.2812	42.94	1280	0.3931
0.2703	43.1	1285	0.3889
0.291	43.27	1290	0.3878
0.2685	43.44	1295	0.3952
0.2924	43.61	1300	0.4158
0.2634	43.77	1305	0.3948
0.2759	43.94	1310	0.3899
0.263	44.11	1315	0.3930
0.2747	44.28	1320	0.4066
0.2842	44.44	1325	0.4001
0.2615	44.61	1330	0.3745
0.2787	44.78	1335	0.3675
0.2945	44.95	1340	0.4024
0.2693	45.12	1345	0.4002
0.2727	45.28	1350	0.3880
0.2724	45.45	1355	0.3729
0.2856	45.62	1360	0.3974
0.2719	45.79	1365	0.3912
0.2842	45.95	1370	0.3626
0.2575	46.12	1375	0.4222
0.2742	46.29	1380	0.3857
0.3015	46.46	1385	0.3772
0.2805	46.62	1390	0.4026
0.2793	46.79	1395	0.3862
0.2735	46.96	1400	0.3896
0.2915	47.13	1405	0.3672
0.2783	47.3	1410	0.3856
0.2799	47.46	1415	0.3663
0.2631	47.63	1420	0.3861
0.261	47.8	1425	0.3760
0.2928	47.97	1430	0.4153
0.2955	48.13	1435	0.3910
0.2751	48.3	1440	0.3880
0.3071	48.47	1445	0.3855
0.2908	48.64	1450	0.3677

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2