glpn-nyu-finetuned-diode-230117-091155

This model is a fine-tuned version of vinvino02/glpn-nyu on the diode-subset dataset. It achieves the following results on the evaluation set:

Loss: 0.4508
Mae: 0.4313
Rmse: 0.6190
Abs Rel: 0.4812
Log Mae: 0.1758
Log Rmse: 0.2320
Delta1: 0.3634
Delta2: 0.6241
Delta3: 0.8095

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: 0.0005
train_batch_size: 24
eval_batch_size: 48
seed: 2022
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.15
num_epochs: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mae	Rmse	Abs Rel	Log Mae	Log Rmse	Delta1	Delta2	Delta3
0.9462	1.0	72	0.4914	0.4683	0.6367	0.5686	0.1952	0.2498	0.3207	0.5284	0.7432
0.4614	2.0	144	0.4642	0.4481	0.6276	0.4985	0.1851	0.2379	0.3310	0.5818	0.7870
0.4575	3.0	216	0.4863	0.4690	0.6352	0.5562	0.1952	0.2474	0.3124	0.5296	0.7341
0.4333	4.0	288	0.4578	0.4350	0.6234	0.4770	0.1772	0.2337	0.3717	0.6145	0.7898
0.461	5.0	360	0.4652	0.4417	0.6251	0.4889	0.1802	0.2351	0.3640	0.5982	0.7674
0.4556	6.0	432	0.4488	0.4268	0.6176	0.4490	0.1727	0.2272	0.3840	0.6282	0.7923
0.4173	7.0	504	0.4583	0.4381	0.6261	0.4704	0.1788	0.2339	0.3806	0.5943	0.7748
0.407	8.0	576	0.4761	0.4597	0.6332	0.5354	0.1911	0.2444	0.3238	0.5456	0.7837
0.4414	9.0	648	0.4753	0.4528	0.6281	0.5253	0.1866	0.2408	0.3403	0.5715	0.7462
0.4027	10.0	720	0.4822	0.4629	0.6333	0.5371	0.1914	0.2438	0.3318	0.5426	0.7352
0.391	11.0	792	0.4488	0.4376	0.6440	0.4148	0.1784	0.2341	0.3712	0.6422	0.7924
0.4323	12.0	864	0.4845	0.4653	0.6343	0.5571	0.1937	0.2474	0.3171	0.5352	0.7550
0.4564	13.0	936	0.4757	0.4606	0.6315	0.5164	0.1901	0.2401	0.3285	0.5452	0.7361
0.4551	14.0	1008	0.4518	0.4354	0.6210	0.4406	0.1770	0.2264	0.3644	0.6064	0.7760
0.4634	15.0	1080	0.4542	0.4306	0.6205	0.4378	0.1756	0.2280	0.3770	0.6093	0.7818
0.4395	16.0	1152	0.4744	0.4438	0.6266	0.5195	0.1828	0.2405	0.3520	0.5851	0.7937
0.4018	17.0	1224	0.4872	0.4567	0.6323	0.5525	0.1888	0.2461	0.3405	0.5610	0.7444
0.4684	18.0	1296	0.4979	0.4736	0.6403	0.5876	0.1977	0.2533	0.3116	0.5289	0.7149
0.408	19.0	1368	0.4995	0.4795	0.6416	0.5995	0.2004	0.2546	0.3013	0.5164	0.6961
0.4372	20.0	1440	0.4580	0.4346	0.6415	0.4046	0.1794	0.2366	0.3786	0.6417	0.7788
0.4104	21.0	1512	0.4781	0.4531	0.6293	0.5330	0.1868	0.2426	0.3454	0.5700	0.7470
0.3843	22.0	1584	0.4487	0.4322	0.6203	0.4518	0.1755	0.2280	0.3711	0.6160	0.7828
0.4025	23.0	1656	0.4659	0.4448	0.6341	0.4721	0.1824	0.2376	0.3563	0.6061	0.7633
0.4553	24.0	1728	0.4766	0.4557	0.6301	0.5319	0.1890	0.2429	0.3306	0.5491	0.7753
0.4134	25.0	1800	0.4779	0.4530	0.6289	0.5299	0.1871	0.2424	0.3416	0.5645	0.7550
0.3913	26.0	1872	0.4591	0.4392	0.6292	0.4691	0.1799	0.2351	0.3610	0.6102	0.7819
0.373	27.0	1944	0.4634	0.4330	0.6313	0.4694	0.1767	0.2376	0.3885	0.6201	0.7928
0.4079	28.0	2016	0.4820	0.4589	0.6308	0.5405	0.1899	0.2444	0.3339	0.5510	0.7570
0.4032	29.0	2088	0.4707	0.4488	0.6405	0.4577	0.1867	0.2428	0.3565	0.5864	0.7626
0.4125	30.0	2160	0.4897	0.4624	0.6376	0.5479	0.1921	0.2481	0.3268	0.5538	0.7390
0.4074	31.0	2232	0.4599	0.4410	0.6300	0.4652	0.1808	0.2350	0.3540	0.6147	0.7724
0.3915	32.0	2304	0.4677	0.4447	0.6275	0.5075	0.1829	0.2390	0.3537	0.5876	0.7824
0.4205	33.0	2376	0.4820	0.4525	0.6414	0.5255	0.1868	0.2475	0.3568	0.5918	0.7502
0.3802	34.0	2448	0.4654	0.4452	0.6293	0.4914	0.1828	0.2372	0.3525	0.5906	0.7701
0.3744	35.0	2520	0.4389	0.4232	0.6263	0.4165	0.1708	0.2273	0.3972	0.6406	0.8019
0.3607	36.0	2592	0.4575	0.4381	0.6200	0.4773	0.1793	0.2316	0.3534	0.5940	0.7894
0.355	37.0	2664	0.4859	0.4649	0.6340	0.5495	0.1929	0.2462	0.3201	0.5454	0.7249
0.3934	38.0	2736	0.4747	0.4518	0.6296	0.5278	0.1866	0.2421	0.3319	0.5776	0.7704
0.3839	39.0	2808	0.4835	0.4571	0.6321	0.5492	0.1887	0.2456	0.3380	0.5675	0.7425
0.4076	40.0	2880	0.4694	0.4506	0.6273	0.5213	0.1859	0.2403	0.3293	0.5790	0.7860
0.386	41.0	2952	0.4809	0.4555	0.6304	0.5455	0.1884	0.2444	0.3383	0.5602	0.7611
0.368	42.0	3024	0.4471	0.4303	0.6174	0.4588	0.1751	0.2283	0.3574	0.6213	0.8128
0.3776	43.0	3096	0.4677	0.4396	0.6268	0.5002	0.1796	0.2379	0.3642	0.6126	0.7733
0.3461	44.0	3168	0.4793	0.4530	0.6297	0.5130	0.1875	0.2432	0.3373	0.5752	0.7565
0.3595	45.0	3240	0.4758	0.4577	0.6298	0.5266	0.1893	0.2413	0.3236	0.5600	0.7542
0.3825	46.0	3312	0.4758	0.4509	0.6361	0.5052	0.1853	0.2429	0.3518	0.5872	0.7610
0.3953	47.0	3384	0.4908	0.4648	0.6372	0.5834	0.1927	0.2505	0.3239	0.5481	0.7420
0.3438	48.0	3456	0.4689	0.4451	0.6277	0.5108	0.1825	0.2390	0.3495	0.5939	0.7826
0.3849	49.0	3528	0.4555	0.4290	0.6234	0.4742	0.1745	0.2335	0.3857	0.6304	0.8015
0.3676	50.0	3600	0.4785	0.4553	0.6338	0.5359	0.1880	0.2447	0.3361	0.5723	0.7647
0.3426	51.0	3672	0.4629	0.4405	0.6268	0.5024	0.1810	0.2383	0.3515	0.6024	0.7976
0.3406	52.0	3744	0.4785	0.4527	0.6305	0.5474	0.1869	0.2448	0.3392	0.5738	0.7701
0.3478	53.0	3816	0.4472	0.4224	0.6203	0.4576	0.1707	0.2305	0.3937	0.6448	0.8120
0.3408	54.0	3888	0.4611	0.4430	0.6244	0.4869	0.1815	0.2344	0.3480	0.5937	0.7742
0.3891	55.0	3960	0.4698	0.4464	0.6289	0.5206	0.1838	0.2406	0.3385	0.5851	0.8102
0.3667	56.0	4032	0.4919	0.4670	0.6359	0.5822	0.1941	0.2505	0.3171	0.5391	0.7431
0.3387	57.0	4104	0.4540	0.4294	0.6269	0.4589	0.1746	0.2331	0.3770	0.6414	0.8027
0.3458	58.0	4176	0.4843	0.4584	0.6329	0.5605	0.1896	0.2468	0.3299	0.5630	0.7564
0.3379	59.0	4248	0.4718	0.4511	0.6292	0.5274	0.1861	0.2414	0.3335	0.5728	0.7886
0.3321	60.0	4320	0.4390	0.4261	0.6234	0.4420	0.1727	0.2289	0.3790	0.6449	0.8035
0.3446	61.0	4392	0.4667	0.4435	0.6256	0.5249	0.1822	0.2397	0.3416	0.5949	0.7975
0.3249	62.0	4464	0.4511	0.4305	0.6228	0.4859	0.1752	0.2342	0.3678	0.6349	0.8087
0.3358	63.0	4536	0.4499	0.4282	0.6235	0.4776	0.1739	0.2335	0.3754	0.6434	0.8118
0.3322	64.0	4608	0.4581	0.4350	0.6231	0.5051	0.1780	0.2367	0.3607	0.6115	0.8097
0.3345	65.0	4680	0.4694	0.4440	0.6312	0.5335	0.1823	0.2428	0.3550	0.5915	0.7989
0.3145	66.0	4752	0.4630	0.4398	0.6249	0.5175	0.1801	0.2385	0.3556	0.6064	0.7975
0.3406	67.0	4824	0.4633	0.4392	0.6233	0.5118	0.1799	0.2376	0.3537	0.6043	0.7952
0.3238	68.0	4896	0.4573	0.4344	0.6222	0.4979	0.1777	0.2354	0.3609	0.6142	0.8109
0.3512	69.0	4968	0.4548	0.4329	0.6235	0.4915	0.1768	0.2352	0.3647	0.6250	0.8067
0.3262	70.0	5040	0.4526	0.4290	0.6215	0.4886	0.1743	0.2338	0.3685	0.6353	0.8157
0.3401	71.0	5112	0.4642	0.4365	0.6225	0.5179	0.1785	0.2381	0.3633	0.6096	0.8025
0.3322	72.0	5184	0.4740	0.4511	0.6275	0.5447	0.1864	0.2432	0.3310	0.5703	0.7911
0.3163	73.0	5256	0.4669	0.4424	0.6260	0.5182	0.1813	0.2391	0.3477	0.5993	0.8008
0.3193	74.0	5328	0.4579	0.4322	0.6258	0.4886	0.1763	0.2361	0.3693	0.6271	0.8030
0.3241	75.0	5400	0.4647	0.4381	0.6226	0.5126	0.1796	0.2376	0.3520	0.6098	0.8003
0.3212	76.0	5472	0.4480	0.4241	0.6201	0.4641	0.1719	0.2307	0.3843	0.6451	0.8112
0.333	77.0	5544	0.4479	0.4259	0.6174	0.4686	0.1730	0.2299	0.3752	0.6358	0.8140
0.3214	78.0	5616	0.4469	0.4249	0.6185	0.4691	0.1723	0.2306	0.3843	0.6376	0.8124
0.3393	79.0	5688	0.4646	0.4411	0.6241	0.5228	0.1810	0.2389	0.3501	0.5960	0.8035
0.3346	80.0	5760	0.4534	0.4315	0.6222	0.4803	0.1760	0.2333	0.3723	0.6189	0.8068
0.3238	81.0	5832	0.4521	0.4305	0.6221	0.4726	0.1754	0.2324	0.3701	0.6238	0.8099
0.3265	82.0	5904	0.4541	0.4335	0.6189	0.4880	0.1773	0.2330	0.3571	0.6134	0.8114
0.3074	83.0	5976	0.4613	0.4383	0.6219	0.5027	0.1797	0.2360	0.3545	0.6016	0.7990
0.317	84.0	6048	0.4513	0.4290	0.6178	0.4768	0.1746	0.2313	0.3718	0.6235	0.8083
0.3363	85.0	6120	0.4508	0.4291	0.6201	0.4744	0.1745	0.2318	0.3733	0.6317	0.8070
0.3159	86.0	6192	0.4604	0.4364	0.6217	0.5065	0.1786	0.2365	0.3581	0.6093	0.8020
0.3275	87.0	6264	0.4489	0.4293	0.6203	0.4706	0.1745	0.2308	0.3682	0.6319	0.8102
0.3102	88.0	6336	0.4484	0.4262	0.6183	0.4763	0.1733	0.2315	0.3765	0.6379	0.8143
0.301	89.0	6408	0.4615	0.4390	0.6216	0.5104	0.1800	0.2368	0.3518	0.6026	0.8002
0.3273	90.0	6480	0.4508	0.4301	0.6198	0.4805	0.1753	0.2325	0.3654	0.6276	0.8123
0.3333	91.0	6552	0.4525	0.4316	0.6202	0.4804	0.1762	0.2325	0.3655	0.6199	0.8077
0.3192	92.0	6624	0.4535	0.4331	0.6196	0.4863	0.1769	0.2330	0.3611	0.6176	0.8066
0.3221	93.0	6696	0.4492	0.4287	0.6180	0.4777	0.1745	0.2313	0.3691	0.6293	0.8117
0.3122	94.0	6768	0.4554	0.4342	0.6199	0.4927	0.1775	0.2339	0.3564	0.6156	0.8091
0.3133	95.0	6840	0.4547	0.4338	0.6197	0.4957	0.1772	0.2342	0.3566	0.6191	0.8096
0.3259	96.0	6912	0.4526	0.4324	0.6197	0.4893	0.1767	0.2334	0.3601	0.6205	0.8112
0.2985	97.0	6984	0.4526	0.4332	0.6199	0.4879	0.1768	0.2332	0.3590	0.6201	0.8085
0.3071	98.0	7056	0.4520	0.4325	0.6197	0.4861	0.1765	0.2329	0.3606	0.6211	0.8098
0.3326	99.0	7128	0.4494	0.4297	0.6183	0.4777	0.1751	0.2314	0.3668	0.6264	0.8106
0.3115	100.0	7200	0.4508	0.4313	0.6190	0.4812	0.1758	0.2320	0.3634	0.6241	0.8095

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-230117-091155

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js