glpn-nyu-finetuned-diode-221230-040136

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: 0.4420
Mae: 0.4258
Rmse: 0.6168
Abs Rel: 0.4569
Log Mae: 0.1725
Log Rmse: 0.2275
Delta1: 0.3673
Delta2: 0.6387
Delta3: 0.8166

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	0.4928	0.4687	0.6431	0.5676	0.1955	0.2515	0.3153	0.5292	0.7843
0.4692	2.0	144	0.4566	0.4433	0.6289	0.4690	0.1822	0.2344	0.3383	0.6035	0.7865
0.4633	3.0	216	0.4811	0.4639	0.6338	0.5407	0.1928	0.2452	0.3191	0.5373	0.7500
0.4369	4.0	288	0.4665	0.4449	0.6268	0.5086	0.1827	0.2387	0.3504	0.5870	0.7847
0.4626	5.0	360	0.4516	0.4332	0.6228	0.4622	0.1759	0.2311	0.3713	0.6231	0.7962
0.4556	6.0	432	0.4443	0.4207	0.6180	0.4486	0.1695	0.2278	0.3989	0.6454	0.8095
0.4104	7.0	504	0.4674	0.4411	0.6250	0.5099	0.1810	0.2384	0.3602	0.5962	0.7832
0.4035	8.0	576	0.4752	0.4547	0.6318	0.5064	0.1874	0.2396	0.3435	0.5580	0.7556
0.4215	9.0	648	0.4879	0.4711	0.6411	0.5626	0.1958	0.2496	0.3184	0.5331	0.7103
0.404	10.0	720	0.4715	0.4479	0.6276	0.5254	0.1841	0.2406	0.3463	0.5859	0.7681
0.3788	11.0	792	0.4607	0.4487	0.6248	0.4814	0.1835	0.2324	0.3458	0.5698	0.7540
0.4399	12.0	864	0.4925	0.4609	0.6363	0.5713	0.1910	0.2499	0.3317	0.5621	0.7471
0.4287	13.0	936	0.4784	0.4463	0.6321	0.5211	0.1836	0.2430	0.3581	0.5865	0.7841
0.4079	14.0	1008	0.4664	0.4514	0.6289	0.4921	0.1848	0.2360	0.3463	0.5705	0.7463
0.4246	15.0	1080	0.4734	0.4548	0.6318	0.5166	0.1872	0.2410	0.3355	0.5666	0.7618
0.4015	16.0	1152	0.4874	0.4581	0.6324	0.5738	0.1894	0.2484	0.3354	0.5611	0.7531
0.3769	17.0	1224	0.4927	0.4622	0.6341	0.5808	0.1911	0.2497	0.3326	0.5531	0.7436
0.4393	18.0	1296	0.4903	0.4666	0.6355	0.5729	0.1937	0.2496	0.3235	0.5415	0.7276
0.3765	19.0	1368	0.4808	0.4568	0.6311	0.5585	0.1889	0.2460	0.3319	0.5596	0.7620
0.394	20.0	1440	0.4363	0.4262	0.6366	0.4018	0.1718	0.2302	0.4056	0.6590	0.7972
0.374	21.0	1512	0.4500	0.4262	0.6286	0.4592	0.1726	0.2339	0.3895	0.6529	0.8082
0.3511	22.0	1584	0.4436	0.4316	0.6251	0.4535	0.1757	0.2300	0.3593	0.6330	0.8068
0.3999	23.0	1656	0.4715	0.4493	0.6285	0.5333	0.1847	0.2414	0.3363	0.5844	0.7834
0.4154	24.0	1728	0.4659	0.4415	0.6248	0.5114	0.1807	0.2379	0.3510	0.5953	0.7975
0.3788	25.0	1800	0.4463	0.4292	0.6274	0.4328	0.1735	0.2298	0.3851	0.6403	0.7899
0.3676	26.0	1872	0.4444	0.4233	0.6160	0.4481	0.1707	0.2264	0.3900	0.6411	0.7991
0.3538	27.0	1944	0.4533	0.4435	0.6234	0.4714	0.1817	0.2311	0.3387	0.5813	0.7915
0.3678	28.0	2016	0.4808	0.4562	0.6302	0.5500	0.1883	0.2445	0.3327	0.5654	0.7515
0.359	29.0	2088	0.4444	0.4287	0.6248	0.4480	0.1734	0.2296	0.3700	0.6498	0.8025
0.3583	30.0	2160	0.4484	0.4295	0.6245	0.4651	0.1743	0.2319	0.3762	0.6359	0.8008
0.3709	31.0	2232	0.4706	0.4528	0.6282	0.5271	0.1866	0.2410	0.3318	0.5666	0.7740
0.3657	32.0	2304	0.4274	0.4143	0.6127	0.4132	0.1657	0.2200	0.4018	0.6562	0.8217
0.3743	33.0	2376	0.4615	0.4396	0.6223	0.5060	0.1796	0.2357	0.3539	0.6018	0.7928
0.3537	34.0	2448	0.4438	0.4311	0.6182	0.4540	0.1748	0.2272	0.3694	0.6183	0.7994
0.3329	35.0	2520	0.4460	0.4308	0.6187	0.4619	0.1749	0.2287	0.3582	0.6300	0.8110
0.3432	36.0	2592	0.4415	0.4256	0.6186	0.4454	0.1723	0.2267	0.3698	0.6384	0.8169
0.3412	37.0	2664	0.4399	0.4270	0.6212	0.4336	0.1726	0.2265	0.3771	0.6398	0.8031
0.3748	38.0	2736	0.4580	0.4430	0.6245	0.4873	0.1814	0.2341	0.3412	0.5946	0.7889
0.3441	39.0	2808	0.4532	0.4359	0.6223	0.4734	0.1774	0.2318	0.3569	0.6182	0.7917
0.3722	40.0	2880	0.4434	0.4276	0.6175	0.4559	0.1732	0.2276	0.3631	0.6346	0.8109
0.357	41.0	2952	0.4566	0.4361	0.6221	0.4858	0.1780	0.2333	0.3487	0.6218	0.8071
0.3495	42.0	3024	0.4372	0.4186	0.6190	0.4290	0.1681	0.2252	0.3910	0.6704	0.8119
0.342	43.0	3096	0.4366	0.4235	0.6177	0.4336	0.1710	0.2245	0.3725	0.6500	0.8100
0.3352	44.0	3168	0.4306	0.4167	0.6189	0.4201	0.1668	0.2231	0.3913	0.6716	0.8174
0.3345	45.0	3240	0.4406	0.4236	0.6185	0.4507	0.1713	0.2277	0.3788	0.6505	0.8102
0.3492	46.0	3312	0.4425	0.4239	0.6146	0.4556	0.1716	0.2266	0.3674	0.6464	0.8169
0.3646	47.0	3384	0.4484	0.4261	0.6201	0.4575	0.1731	0.2308	0.3739	0.6419	0.8054
0.3246	48.0	3456	0.4526	0.4370	0.6202	0.4814	0.1788	0.2317	0.3450	0.6054	0.8089
0.3566	49.0	3528	0.4469	0.4318	0.6213	0.4606	0.1754	0.2295	0.3587	0.6293	0.8066
0.3401	50.0	3600	0.4525	0.4344	0.6198	0.4872	0.1773	0.2328	0.3519	0.6136	0.8098
0.3284	51.0	3672	0.4418	0.4248	0.6200	0.4468	0.1717	0.2274	0.3742	0.6474	0.8114
0.3252	52.0	3744	0.4366	0.4260	0.6237	0.4292	0.1720	0.2258	0.3681	0.6544	0.8132
0.3255	53.0	3816	0.4579	0.4397	0.6222	0.4924	0.1796	0.2339	0.3487	0.5974	0.8000
0.3158	54.0	3888	0.4530	0.4353	0.6197	0.4875	0.1776	0.2326	0.3499	0.6134	0.8070
0.3257	55.0	3960	0.4335	0.4197	0.6170	0.4365	0.1690	0.2246	0.3847	0.6578	0.8149
0.3217	56.0	4032	0.4488	0.4332	0.6199	0.4697	0.1764	0.2302	0.3560	0.6198	0.8049
0.3232	57.0	4104	0.4392	0.4257	0.6168	0.4456	0.1723	0.2255	0.3669	0.6382	0.8120
0.3259	58.0	4176	0.4371	0.4204	0.6177	0.4362	0.1695	0.2251	0.3906	0.6495	0.8097
0.3226	59.0	4248	0.4475	0.4299	0.6191	0.4712	0.1751	0.2303	0.3561	0.6316	0.8131
0.3127	60.0	4320	0.4405	0.4254	0.6164	0.4518	0.1724	0.2267	0.3727	0.6341	0.8158
0.3168	61.0	4392	0.4491	0.4318	0.6180	0.4722	0.1757	0.2299	0.3606	0.6146	0.8128
0.3059	62.0	4464	0.4432	0.4259	0.6185	0.4597	0.1726	0.2288	0.3700	0.6441	0.8136
0.3043	63.0	4536	0.4373	0.4196	0.6174	0.4398	0.1691	0.2256	0.3817	0.6669	0.8178
0.3143	64.0	4608	0.4500	0.4317	0.6186	0.4774	0.1758	0.2309	0.3559	0.6220	0.8175
0.3084	65.0	4680	0.4431	0.4257	0.6173	0.4600	0.1727	0.2283	0.3666	0.6427	0.8170
0.3034	66.0	4752	0.4472	0.4295	0.6183	0.4719	0.1745	0.2301	0.3627	0.6297	0.8112
0.3337	67.0	4824	0.4462	0.4293	0.6179	0.4670	0.1744	0.2292	0.3606	0.6301	0.8130
0.3168	68.0	4896	0.4483	0.4305	0.6182	0.4699	0.1751	0.2298	0.3611	0.6220	0.8114
0.3159	69.0	4968	0.4409	0.4251	0.6166	0.4523	0.1723	0.2269	0.3666	0.6415	0.8155
0.3137	70.0	5040	0.4446	0.4276	0.6179	0.4598	0.1732	0.2283	0.3682	0.6330	0.8110
0.3274	71.0	5112	0.4428	0.4270	0.6176	0.4585	0.1729	0.2279	0.3673	0.6351	0.8162
0.323	72.0	5184	0.4431	0.4264	0.6168	0.4590	0.1729	0.2279	0.3658	0.6362	0.8159
0.3106	73.0	5256	0.4424	0.4256	0.6173	0.4557	0.1722	0.2275	0.3690	0.6403	0.8169
0.3176	74.0	5328	0.4422	0.4255	0.6167	0.4565	0.1725	0.2275	0.3676	0.6397	0.8163
0.3157	75.0	5400	0.4420	0.4258	0.6168	0.4569	0.1725	0.2275	0.3673	0.6387	0.8166

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-221230-040136

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js