glpn-nyu-finetuned-diode-230113-130735

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: 0.4320
Mae: 0.4213
Rmse: 0.6133
Abs Rel: 0.4298
Log Mae: 0.1697
Log Rmse: 0.2216
Delta1: 0.3800
Delta2: 0.6396
Delta3: 0.8189

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: 0.0003
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: 75
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mae	Rmse	Abs Rel	Log Mae	Log Rmse	Delta1	Delta2	Delta3
1.0073	1.0	72	0.4927	0.4684	0.6425	0.5680	0.1955	0.2515	0.3154	0.5289	0.7834
0.4694	2.0	144	0.4560	0.4425	0.6285	0.4674	0.1818	0.2341	0.3395	0.6061	0.7873
0.4632	3.0	216	0.4817	0.4646	0.6341	0.5412	0.1930	0.2453	0.3181	0.5368	0.7491
0.4363	4.0	288	0.4589	0.4379	0.6228	0.4880	0.1793	0.2348	0.3588	0.6025	0.7952
0.4636	5.0	360	0.4767	0.4545	0.6301	0.5367	0.1878	0.2430	0.3279	0.5716	0.7705
0.4642	6.0	432	0.4437	0.4185	0.6200	0.4405	0.1689	0.2283	0.4071	0.6531	0.8091
0.409	7.0	504	0.4787	0.4542	0.6291	0.5399	0.1873	0.2430	0.3345	0.5679	0.7648
0.4081	8.0	576	0.4545	0.4359	0.6258	0.4554	0.1779	0.2311	0.3717	0.6035	0.7952
0.4146	9.0	648	0.4726	0.4523	0.6293	0.5108	0.1870	0.2403	0.3394	0.5692	0.7571
0.392	10.0	720	0.4643	0.4453	0.6249	0.5081	0.1831	0.2372	0.3380	0.5881	0.7917
0.3722	11.0	792	0.4670	0.4475	0.6245	0.4957	0.1838	0.2355	0.3413	0.5739	0.7689
0.4397	12.0	864	0.4548	0.4367	0.6262	0.4604	0.1780	0.2319	0.3664	0.6081	0.7903
0.43	13.0	936	0.4281	0.4223	0.6230	0.3974	0.1691	0.2207	0.3975	0.6426	0.7943
0.3976	14.0	1008	0.4592	0.4470	0.6249	0.4759	0.1827	0.2321	0.3482	0.5784	0.7507
0.4251	15.0	1080	0.4515	0.4366	0.6205	0.4589	0.1773	0.2285	0.3689	0.5990	0.7785
0.4007	16.0	1152	0.4859	0.4668	0.6347	0.5570	0.1939	0.2467	0.3156	0.5378	0.7265
0.376	17.0	1224	0.4529	0.4331	0.6195	0.4421	0.1752	0.2260	0.3795	0.6016	0.7702
0.4028	18.0	1296	0.5027	0.4775	0.6420	0.6169	0.1993	0.2569	0.3098	0.5228	0.7035
0.3816	19.0	1368	0.4869	0.4634	0.6342	0.5565	0.1924	0.2473	0.3276	0.5448	0.7370
0.4092	20.0	1440	0.4317	0.4155	0.6164	0.4083	0.1661	0.2218	0.4003	0.6569	0.8123
0.3673	21.0	1512	0.4433	0.4326	0.6208	0.4295	0.1750	0.2244	0.3751	0.6068	0.7879
0.3698	22.0	1584	0.4607	0.4322	0.6216	0.4981	0.1758	0.2354	0.3831	0.6163	0.7906
0.3771	23.0	1656	0.4668	0.4478	0.6255	0.5075	0.1841	0.2373	0.3390	0.5819	0.7697
0.4343	24.0	1728	0.4532	0.4331	0.6203	0.4722	0.1767	0.2312	0.3587	0.6166	0.8087
0.4011	25.0	1800	0.4499	0.4327	0.6213	0.4519	0.1755	0.2279	0.3716	0.6152	0.7844
0.3714	26.0	1872	0.4460	0.4254	0.6188	0.4495	0.1716	0.2278	0.3932	0.6352	0.7916
0.3436	27.0	1944	0.4360	0.4182	0.6165	0.4192	0.1682	0.2224	0.3894	0.6524	0.8145
0.3698	28.0	2016	0.4694	0.4536	0.6274	0.5040	0.1863	0.2369	0.3356	0.5667	0.7469
0.365	29.0	2088	0.4288	0.4139	0.6156	0.4025	0.1655	0.2199	0.4028	0.6623	0.8109
0.3723	30.0	2160	0.4337	0.4148	0.6141	0.4192	0.1661	0.2215	0.4044	0.6578	0.8073
0.365	31.0	2232	0.4529	0.4309	0.6192	0.4751	0.1755	0.2314	0.3770	0.6115	0.7909
0.3571	32.0	2304	0.4302	0.4151	0.6170	0.4134	0.1663	0.2227	0.4089	0.6611	0.8078
0.3727	33.0	2376	0.4599	0.4352	0.6214	0.4937	0.1776	0.2348	0.3659	0.6120	0.7949
0.3538	34.0	2448	0.4391	0.4257	0.6161	0.4404	0.1720	0.2248	0.3768	0.6317	0.8042
0.3306	35.0	2520	0.4393	0.4223	0.6198	0.4328	0.1702	0.2262	0.3886	0.6493	0.8062
0.3369	36.0	2592	0.4496	0.4316	0.6182	0.4642	0.1751	0.2289	0.3712	0.6124	0.8005
0.3389	37.0	2664	0.4573	0.4376	0.6213	0.4897	0.1787	0.2338	0.3628	0.6014	0.7932
0.3767	38.0	2736	0.4558	0.4366	0.6216	0.4840	0.1786	0.2334	0.3566	0.6064	0.7973
0.3462	39.0	2808	0.4580	0.4380	0.6221	0.4815	0.1785	0.2328	0.3640	0.6020	0.7850
0.3834	40.0	2880	0.4664	0.4459	0.6245	0.5155	0.1836	0.2385	0.3426	0.5782	0.7944
0.3564	41.0	2952	0.4452	0.4271	0.6175	0.4563	0.1733	0.2282	0.3749	0.6269	0.8081
0.3571	42.0	3024	0.4357	0.4189	0.6151	0.4360	0.1686	0.2243	0.3947	0.6482	0.8163
0.345	43.0	3096	0.4285	0.4130	0.6114	0.4173	0.1653	0.2202	0.4034	0.6611	0.8223
0.3163	44.0	3168	0.4473	0.4274	0.6176	0.4624	0.1732	0.2288	0.3790	0.6245	0.8095
0.3331	45.0	3240	0.4392	0.4214	0.6139	0.4429	0.1699	0.2244	0.3887	0.6388	0.8081
0.3574	46.0	3312	0.4487	0.4230	0.6156	0.4608	0.1710	0.2282	0.3860	0.6431	0.8063
0.3703	47.0	3384	0.4342	0.4176	0.6179	0.4286	0.1678	0.2247	0.3918	0.6668	0.8098
0.325	48.0	3456	0.4390	0.4238	0.6150	0.4500	0.1715	0.2256	0.3695	0.6334	0.8216
0.3494	49.0	3528	0.4364	0.4182	0.6165	0.4348	0.1680	0.2248	0.4041	0.6539	0.8104
0.3439	50.0	3600	0.4401	0.4252	0.6156	0.4414	0.1716	0.2243	0.3831	0.6260	0.8042
0.3235	51.0	3672	0.4459	0.4258	0.6173	0.4607	0.1728	0.2287	0.3819	0.6272	0.8106
0.3197	52.0	3744	0.4341	0.4205	0.6153	0.4291	0.1691	0.2226	0.3874	0.6429	0.8173
0.3231	53.0	3816	0.4499	0.4297	0.6180	0.4654	0.1745	0.2290	0.3730	0.6166	0.8053
0.3182	54.0	3888	0.4407	0.4242	0.6145	0.4501	0.1714	0.2252	0.3762	0.6366	0.8124
0.334	55.0	3960	0.4518	0.4335	0.6176	0.4773	0.1768	0.2304	0.3591	0.6065	0.8111
0.3198	56.0	4032	0.4505	0.4322	0.6173	0.4725	0.1760	0.2298	0.3637	0.6131	0.8025
0.3165	57.0	4104	0.4378	0.4248	0.6174	0.4369	0.1720	0.2246	0.3729	0.6377	0.8137
0.3269	58.0	4176	0.4372	0.4275	0.6156	0.4415	0.1730	0.2240	0.3675	0.6276	0.8095
0.3224	59.0	4248	0.4359	0.4244	0.6149	0.4351	0.1711	0.2231	0.3721	0.6366	0.8090
0.3104	60.0	4320	0.4317	0.4209	0.6146	0.4284	0.1696	0.2220	0.3799	0.6395	0.8179
0.3248	61.0	4392	0.4323	0.4207	0.6138	0.4268	0.1694	0.2216	0.3864	0.6386	0.8148
0.303	62.0	4464	0.4309	0.4189	0.6126	0.4264	0.1685	0.2213	0.3853	0.6453	0.8194
0.3126	63.0	4536	0.4308	0.4206	0.6141	0.4229	0.1693	0.2208	0.3783	0.6447	0.8162
0.3099	64.0	4608	0.4330	0.4239	0.6149	0.4298	0.1709	0.2218	0.3709	0.6323	0.8182
0.3075	65.0	4680	0.4322	0.4222	0.6144	0.4276	0.1701	0.2217	0.3784	0.6374	0.8159
0.3024	66.0	4752	0.4393	0.4269	0.6155	0.4456	0.1729	0.2249	0.3722	0.6245	0.8100
0.3319	67.0	4824	0.4385	0.4273	0.6155	0.4402	0.1728	0.2238	0.3722	0.6244	0.8085
0.3163	68.0	4896	0.4334	0.4215	0.6128	0.4305	0.1699	0.2216	0.3814	0.6379	0.8145
0.3219	69.0	4968	0.4298	0.4197	0.6131	0.4215	0.1688	0.2203	0.3821	0.6453	0.8170
0.3155	70.0	5040	0.4295	0.4199	0.6134	0.4219	0.1687	0.2204	0.3846	0.6453	0.8164
0.3265	71.0	5112	0.4294	0.4194	0.6123	0.4232	0.1687	0.2203	0.3804	0.6468	0.8203
0.3231	72.0	5184	0.4338	0.4231	0.6138	0.4333	0.1707	0.2222	0.3775	0.6340	0.8166
0.3077	73.0	5256	0.4327	0.4221	0.6134	0.4315	0.1702	0.2219	0.3800	0.6361	0.8185
0.3178	74.0	5328	0.4312	0.4203	0.6126	0.4278	0.1693	0.2212	0.3813	0.6417	0.8194
0.3157	75.0	5400	0.4320	0.4213	0.6133	0.4298	0.1697	0.2216	0.3800	0.6396	0.8189

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-230113-130735

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js