glpn-nyu-finetuned-diode-230124-104649

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: 0.4340
Mae: 0.4201
Rmse: 0.6110
Abs Rel: 0.4400
Log Mae: 0.1698
Log Rmse: 0.2229
Delta1: 0.3745
Delta2: 0.6423
Delta3: 0.8241

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: 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
1.0761	1.0	72	0.5035	0.4784	0.6697	0.5506	0.2007	0.2592	0.3019	0.5331	0.7997
0.4776	2.0	144	0.4640	0.4494	0.6305	0.4846	0.1853	0.2370	0.3321	0.5850	0.7752
0.4667	3.0	216	0.4852	0.4716	0.6377	0.5477	0.1965	0.2473	0.3105	0.5246	0.7218
0.4387	4.0	288	0.4587	0.4378	0.6223	0.4874	0.1790	0.2343	0.3577	0.6064	0.7906
0.4612	5.0	360	0.4843	0.4610	0.6341	0.5444	0.1906	0.2458	0.3269	0.5602	0.7445
0.4564	6.0	432	0.4605	0.4330	0.6197	0.4901	0.1764	0.2339	0.3775	0.6049	0.8022
0.4166	7.0	504	0.4576	0.4421	0.6256	0.4625	0.1809	0.2322	0.3613	0.5882	0.7685
0.3922	8.0	576	0.4805	0.4537	0.6296	0.5422	0.1875	0.2439	0.3381	0.5612	0.7874
0.3944	9.0	648	0.4601	0.4430	0.6254	0.4762	0.1812	0.2332	0.3545	0.5877	0.7662
0.3748	10.0	720	0.4606	0.4377	0.6221	0.4960	0.1796	0.2354	0.3573	0.5960	0.8031
0.3749	11.0	792	0.4513	0.4377	0.6300	0.4403	0.1786	0.2311	0.3621	0.6083	0.7901
0.4259	12.0	864	0.4834	0.4519	0.6328	0.5462	0.1862	0.2457	0.3521	0.5777	0.7614
0.4337	13.0	936	0.4338	0.4153	0.6214	0.4096	0.1664	0.2248	0.4137	0.6651	0.8037
0.4032	14.0	1008	0.4640	0.4544	0.6279	0.4922	0.1868	0.2351	0.3286	0.5574	0.7557
0.4286	15.0	1080	0.4835	0.4651	0.6338	0.5567	0.1929	0.2465	0.3197	0.5449	0.7304
0.397	16.0	1152	0.4713	0.4547	0.6279	0.5121	0.1872	0.2383	0.3312	0.5644	0.7456
0.3713	17.0	1224	0.4664	0.4375	0.6290	0.4766	0.1780	0.2361	0.3821	0.6036	0.7668
0.4155	18.0	1296	0.4881	0.4722	0.6367	0.5705	0.1967	0.2494	0.3061	0.5293	0.7220
0.3822	19.0	1368	0.4819	0.4592	0.6322	0.5526	0.1898	0.2456	0.3357	0.5531	0.7365
0.408	20.0	1440	0.4367	0.4201	0.6141	0.4322	0.1691	0.2232	0.3902	0.6418	0.8083
0.3698	21.0	1512	0.4461	0.4263	0.6171	0.4454	0.1724	0.2263	0.3850	0.6232	0.7980
0.3628	22.0	1584	0.4461	0.4254	0.6226	0.4520	0.1724	0.2307	0.3953	0.6324	0.7926
0.3827	23.0	1656	0.4753	0.4529	0.6364	0.5288	0.1867	0.2450	0.3415	0.5893	0.7678
0.4378	24.0	1728	0.4779	0.4608	0.6308	0.5422	0.1907	0.2433	0.3247	0.5435	0.7434
0.3766	25.0	1800	0.4533	0.4415	0.6231	0.4750	0.1802	0.2309	0.3544	0.5796	0.7917
0.3642	26.0	1872	0.4520	0.4276	0.6224	0.4686	0.1736	0.2322	0.3901	0.6242	0.8048
0.3503	27.0	1944	0.4451	0.4262	0.6163	0.4574	0.1730	0.2278	0.3721	0.6306	0.8116
0.3723	28.0	2016	0.4617	0.4451	0.6239	0.4936	0.1824	0.2346	0.3436	0.5864	0.7740
0.3739	29.0	2088	0.4468	0.4295	0.6209	0.4513	0.1741	0.2285	0.3738	0.6288	0.7954
0.3699	30.0	2160	0.4494	0.4334	0.6233	0.4682	0.1766	0.2320	0.3684	0.6155	0.7947
0.3573	31.0	2232	0.4603	0.4385	0.6215	0.4963	0.1793	0.2345	0.3620	0.5948	0.7839
0.3684	32.0	2304	0.4488	0.4278	0.6195	0.4571	0.1735	0.2290	0.3895	0.6201	0.7970
0.3911	33.0	2376	0.4499	0.4309	0.6201	0.4636	0.1751	0.2301	0.3839	0.6118	0.7803
0.3416	34.0	2448	0.4515	0.4298	0.6185	0.4734	0.1748	0.2311	0.3824	0.6152	0.7916
0.3345	35.0	2520	0.4434	0.4247	0.6163	0.4548	0.1720	0.2274	0.3881	0.6233	0.8077
0.3436	36.0	2592	0.4561	0.4370	0.6208	0.4926	0.1785	0.2337	0.3586	0.5960	0.7979
0.3411	37.0	2664	0.4805	0.4629	0.6337	0.5600	0.1920	0.2468	0.3187	0.5448	0.7601
0.3755	38.0	2736	0.4566	0.4365	0.6235	0.4780	0.1784	0.2335	0.3662	0.5972	0.7941
0.3456	39.0	2808	0.4665	0.4500	0.6259	0.5163	0.1851	0.2386	0.3368	0.5756	0.7686
0.3829	40.0	2880	0.4720	0.4527	0.6279	0.5323	0.1871	0.2423	0.3384	0.5656	0.7635
0.3645	41.0	2952	0.4380	0.4211	0.6133	0.4377	0.1701	0.2234	0.3945	0.6275	0.8056
0.3654	42.0	3024	0.4228	0.4087	0.6240	0.3844	0.1624	0.2220	0.4339	0.6953	0.8065
0.3694	43.0	3096	0.4390	0.4183	0.6153	0.4374	0.1683	0.2250	0.3991	0.6509	0.8065
0.329	44.0	3168	0.4559	0.4349	0.6191	0.4912	0.1775	0.2330	0.3611	0.6075	0.7981
0.3509	45.0	3240	0.4566	0.4341	0.6202	0.4973	0.1774	0.2347	0.3653	0.6168	0.7942
0.3666	46.0	3312	0.4665	0.4452	0.6239	0.5179	0.1830	0.2379	0.3384	0.5860	0.7844
0.3948	47.0	3384	0.4570	0.4406	0.6221	0.4883	0.1805	0.2333	0.3504	0.5887	0.7961
0.3349	48.0	3456	0.4539	0.4372	0.6186	0.4851	0.1789	0.2316	0.3467	0.5966	0.8092
0.3689	49.0	3528	0.4416	0.4182	0.6136	0.4565	0.1685	0.2270	0.3991	0.6475	0.8157
0.3477	50.0	3600	0.4417	0.4241	0.6184	0.4513	0.1713	0.2272	0.3802	0.6461	0.8114
0.3476	51.0	3672	0.4502	0.4333	0.6189	0.4766	0.1763	0.2304	0.3594	0.6120	0.8096
0.3318	52.0	3744	0.4480	0.4268	0.6167	0.4666	0.1728	0.2287	0.3744	0.6318	0.8080
0.336	53.0	3816	0.4504	0.4266	0.6159	0.4792	0.1730	0.2306	0.3782	0.6248	0.8089
0.3283	54.0	3888	0.4490	0.4265	0.6184	0.4689	0.1732	0.2305	0.3872	0.6295	0.8037
0.3465	55.0	3960	0.4371	0.4216	0.6189	0.4399	0.1701	0.2263	0.3866	0.6515	0.8168
0.3299	56.0	4032	0.4544	0.4377	0.6199	0.4828	0.1787	0.2319	0.3532	0.6004	0.7961
0.3301	57.0	4104	0.4351	0.4208	0.6151	0.4317	0.1700	0.2234	0.3837	0.6386	0.8147
0.3314	58.0	4176	0.4347	0.4189	0.6130	0.4373	0.1689	0.2234	0.3889	0.6468	0.8153
0.328	59.0	4248	0.4536	0.4342	0.6187	0.4887	0.1773	0.2326	0.3554	0.6080	0.8052
0.3153	60.0	4320	0.4393	0.4206	0.6130	0.4515	0.1699	0.2259	0.3854	0.6416	0.8156
0.3274	61.0	4392	0.4482	0.4275	0.6148	0.4738	0.1740	0.2295	0.3703	0.6176	0.8177
0.3123	62.0	4464	0.4380	0.4172	0.6139	0.4461	0.1678	0.2259	0.4007	0.6569	0.8189
0.3269	63.0	4536	0.4395	0.4186	0.6123	0.4574	0.1690	0.2267	0.3881	0.6507	0.8179
0.3214	64.0	4608	0.4400	0.4229	0.6128	0.4580	0.1714	0.2264	0.3709	0.6391	0.8222
0.3139	65.0	4680	0.4506	0.4295	0.6169	0.4828	0.1748	0.2315	0.3662	0.6223	0.8150
0.306	66.0	4752	0.4391	0.4210	0.6134	0.4565	0.1702	0.2266	0.3802	0.6481	0.8169
0.3375	67.0	4824	0.4511	0.4304	0.6177	0.4807	0.1751	0.2314	0.3683	0.6189	0.8063
0.3199	68.0	4896	0.4409	0.4230	0.6157	0.4615	0.1716	0.2284	0.3796	0.6425	0.8184
0.3286	69.0	4968	0.4424	0.4242	0.6141	0.4608	0.1721	0.2274	0.3752	0.6317	0.8149
0.3168	70.0	5040	0.4250	0.4130	0.6118	0.4139	0.1653	0.2191	0.3987	0.6650	0.8192
0.3316	71.0	5112	0.4391	0.4222	0.6146	0.4486	0.1707	0.2254	0.3795	0.6376	0.8176
0.3305	72.0	5184	0.4455	0.4273	0.6157	0.4623	0.1738	0.2280	0.3743	0.6173	0.8119
0.3135	73.0	5256	0.4407	0.4254	0.6159	0.4513	0.1726	0.2264	0.3688	0.6342	0.8149
0.3364	74.0	5328	0.4421	0.4268	0.6152	0.4561	0.1730	0.2266	0.3675	0.6234	0.8135
0.3188	75.0	5400	0.4480	0.4317	0.6162	0.4746	0.1760	0.2296	0.3555	0.6132	0.8125
0.3125	76.0	5472	0.4346	0.4197	0.6120	0.4389	0.1693	0.2230	0.3802	0.6449	0.8225
0.3179	77.0	5544	0.4437	0.4274	0.6153	0.4633	0.1737	0.2279	0.3686	0.6240	0.8158
0.317	78.0	5616	0.4364	0.4207	0.6127	0.4491	0.1699	0.2250	0.3743	0.6491	0.8237
0.3303	79.0	5688	0.4464	0.4286	0.6172	0.4742	0.1746	0.2304	0.3679	0.6182	0.8206
0.3267	80.0	5760	0.4295	0.4147	0.6099	0.4243	0.1666	0.2201	0.3934	0.6519	0.8233
0.3219	81.0	5832	0.4306	0.4144	0.6101	0.4278	0.1666	0.2209	0.3897	0.6589	0.8240
0.3271	82.0	5904	0.4378	0.4215	0.6125	0.4465	0.1704	0.2246	0.3787	0.6395	0.8198
0.2986	83.0	5976	0.4401	0.4253	0.6136	0.4511	0.1724	0.2254	0.3697	0.6270	0.8186
0.3153	84.0	6048	0.4355	0.4199	0.6111	0.4418	0.1698	0.2232	0.3781	0.6388	0.8250
0.323	85.0	6120	0.4420	0.4262	0.6135	0.4556	0.1731	0.2260	0.3640	0.6235	0.8207
0.308	86.0	6192	0.4359	0.4206	0.6123	0.4421	0.1701	0.2238	0.3774	0.6409	0.8232
0.3076	87.0	6264	0.4329	0.4185	0.6105	0.4347	0.1688	0.2219	0.3791	0.6471	0.8242
0.3089	88.0	6336	0.4256	0.4117	0.6083	0.4180	0.1651	0.2189	0.3949	0.6666	0.8253
0.299	89.0	6408	0.4449	0.4300	0.6152	0.4602	0.1749	0.2270	0.3596	0.6151	0.8156
0.3211	90.0	6480	0.4330	0.4191	0.6106	0.4339	0.1692	0.2218	0.3785	0.6422	0.8252
0.323	91.0	6552	0.4310	0.4167	0.6098	0.4301	0.1680	0.2211	0.3826	0.6508	0.8260
0.3108	92.0	6624	0.4402	0.4259	0.6130	0.4519	0.1730	0.2251	0.3662	0.6260	0.8188
0.3201	93.0	6696	0.4300	0.4166	0.6097	0.4312	0.1679	0.2211	0.3834	0.6512	0.8245
0.3072	94.0	6768	0.4344	0.4217	0.6117	0.4400	0.1706	0.2228	0.3726	0.6356	0.8239
0.3079	95.0	6840	0.4369	0.4236	0.6121	0.4454	0.1716	0.2238	0.3678	0.6308	0.8241
0.3192	96.0	6912	0.4328	0.4189	0.6105	0.4362	0.1691	0.2220	0.3774	0.6441	0.8245
0.2959	97.0	6984	0.4340	0.4203	0.6110	0.4399	0.1700	0.2228	0.3741	0.6409	0.8245
0.3061	98.0	7056	0.4352	0.4208	0.6112	0.4427	0.1703	0.2234	0.3728	0.6402	0.8249
0.3294	99.0	7128	0.4329	0.4191	0.6107	0.4372	0.1693	0.2223	0.3762	0.6451	0.8238
0.3087	100.0	7200	0.4340	0.4201	0.6110	0.4400	0.1698	0.2229	0.3745	0.6423	0.8241

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-230124-104649

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js