glpn-nyu-finetuned-diode-230124-035129

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: 0.4346
Mae: 0.4251
Rmse: 0.6137
Abs Rel: 0.4412
Log Mae: 0.1720
Log Rmse: 0.2234
Delta1: 0.3728
Delta2: 0.6287
Delta3: 0.8112

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	0.5029	0.4778	0.6689	0.5501	0.2005	0.2591	0.3026	0.5337	0.8000
0.4775	2.0	144	0.4651	0.4519	0.6313	0.4910	0.1865	0.2380	0.3286	0.5789	0.7731
0.4669	3.0	216	0.4843	0.4710	0.6374	0.5449	0.1960	0.2468	0.3114	0.5264	0.7239
0.4386	4.0	288	0.4544	0.4342	0.6217	0.4763	0.1771	0.2324	0.3629	0.6167	0.7950
0.4625	5.0	360	0.4854	0.4649	0.6352	0.5522	0.1930	0.2471	0.3184	0.5485	0.7388
0.4519	6.0	432	0.4507	0.4284	0.6178	0.4597	0.1735	0.2288	0.3860	0.6189	0.7960
0.4145	7.0	504	0.4682	0.4523	0.6294	0.4984	0.1860	0.2374	0.3391	0.5647	0.7637
0.3833	8.0	576	0.4391	0.4221	0.6350	0.4043	0.1700	0.2297	0.4133	0.6641	0.7936
0.3944	9.0	648	0.4482	0.4346	0.6323	0.4269	0.1766	0.2314	0.3877	0.6193	0.7755
0.3819	10.0	720	0.4483	0.4301	0.6208	0.4636	0.1748	0.2293	0.3600	0.6339	0.8145
0.3642	11.0	792	0.4370	0.4224	0.6170	0.4214	0.1701	0.2226	0.3825	0.6312	0.8171
0.4094	12.0	864	0.4779	0.4589	0.6330	0.5249	0.1895	0.2422	0.3379	0.5536	0.7308
0.4376	13.0	936	0.4539	0.4334	0.6202	0.4624	0.1762	0.2291	0.3655	0.6114	0.7897
0.4048	14.0	1008	0.4728	0.4522	0.6275	0.5082	0.1854	0.2376	0.3485	0.5644	0.7391
0.4478	15.0	1080	0.4573	0.4336	0.6215	0.4590	0.1761	0.2301	0.3697	0.6090	0.7884
0.4059	16.0	1152	0.4721	0.4424	0.6251	0.5171	0.1815	0.2391	0.3541	0.5930	0.7832
0.3803	17.0	1224	0.4518	0.4242	0.6216	0.4496	0.1709	0.2294	0.3963	0.6389	0.8061
0.4208	18.0	1296	0.4755	0.4583	0.6307	0.5242	0.1893	0.2417	0.3320	0.5575	0.7438
0.3695	19.0	1368	0.4758	0.4462	0.6272	0.5255	0.1836	0.2413	0.3534	0.5852	0.7696
0.4121	20.0	1440	0.4353	0.4227	0.6271	0.4017	0.1703	0.2245	0.3893	0.6499	0.8077
0.3739	21.0	1512	0.4378	0.4156	0.6159	0.4169	0.1666	0.2233	0.4172	0.6515	0.7963
0.3729	22.0	1584	0.4375	0.4218	0.6158	0.4229	0.1696	0.2222	0.3906	0.6356	0.8061
0.3921	23.0	1656	0.4353	0.4167	0.6141	0.4192	0.1670	0.2218	0.4059	0.6504	0.8023
0.4527	24.0	1728	0.4850	0.4528	0.6307	0.5521	0.1873	0.2459	0.3459	0.5707	0.7571
0.4064	25.0	1800	0.4446	0.4276	0.6172	0.4488	0.1731	0.2263	0.3764	0.6246	0.8056
0.3827	26.0	1872	0.4476	0.4299	0.6199	0.4586	0.1746	0.2293	0.3694	0.6303	0.7972
0.3416	27.0	1944	0.4428	0.4298	0.6228	0.4408	0.1743	0.2272	0.3724	0.6274	0.8103
0.3774	28.0	2016	0.4554	0.4373	0.6206	0.4804	0.1784	0.2321	0.3589	0.6068	0.7875
0.3577	29.0	2088	0.4593	0.4445	0.6232	0.4824	0.1820	0.2329	0.3422	0.5875	0.7750
0.3607	30.0	2160	0.4532	0.4287	0.6219	0.4736	0.1741	0.2330	0.3873	0.6265	0.7896
0.3902	31.0	2232	0.4706	0.4474	0.6257	0.5194	0.1838	0.2394	0.3480	0.5840	0.7677
0.3728	32.0	2304	0.4429	0.4268	0.6184	0.4551	0.1734	0.2280	0.3698	0.6389	0.8076
0.3922	33.0	2376	0.4408	0.4253	0.6161	0.4409	0.1718	0.2247	0.3841	0.6260	0.7934
0.3559	34.0	2448	0.4316	0.4169	0.6126	0.4216	0.1673	0.2211	0.4017	0.6455	0.8065
0.3416	35.0	2520	0.4352	0.4188	0.6138	0.4193	0.1683	0.2211	0.3865	0.6476	0.8133
0.3446	36.0	2592	0.4540	0.4353	0.6208	0.4751	0.1766	0.2307	0.3646	0.6104	0.7931
0.345	37.0	2664	0.4488	0.4299	0.6186	0.4637	0.1741	0.2292	0.3780	0.6227	0.7921
0.396	38.0	2736	0.4405	0.4274	0.6183	0.4471	0.1728	0.2263	0.3733	0.6365	0.8010
0.3816	39.0	2808	0.4683	0.4496	0.6263	0.5104	0.1847	0.2380	0.3439	0.5750	0.7568
0.3976	40.0	2880	0.4462	0.4281	0.6168	0.4603	0.1737	0.2278	0.3733	0.6263	0.8022
0.3645	41.0	2952	0.4474	0.4298	0.6188	0.4630	0.1748	0.2294	0.3744	0.6144	0.8064
0.3525	42.0	3024	0.4262	0.4105	0.6135	0.4148	0.1640	0.2212	0.4184	0.6707	0.8218
0.3615	43.0	3096	0.4251	0.4058	0.6101	0.4083	0.1613	0.2193	0.4230	0.6782	0.8171
0.3358	44.0	3168	0.4594	0.4411	0.6216	0.4932	0.1804	0.2340	0.3528	0.5940	0.7864
0.3379	45.0	3240	0.4454	0.4267	0.6205	0.4642	0.1727	0.2300	0.3827	0.6364	0.8102
0.367	46.0	3312	0.4631	0.4371	0.6211	0.5096	0.1787	0.2365	0.3648	0.6074	0.7929
0.3814	47.0	3384	0.4686	0.4450	0.6259	0.5316	0.1827	0.2406	0.3480	0.5939	0.7808
0.3374	48.0	3456	0.4383	0.4154	0.6182	0.4298	0.1666	0.2257	0.4200	0.6578	0.8079
0.3789	49.0	3528	0.4549	0.4337	0.6216	0.4961	0.1765	0.2344	0.3712	0.6160	0.7987
0.3513	50.0	3600	0.4550	0.4363	0.6235	0.4865	0.1777	0.2334	0.3648	0.6111	0.7886
0.3355	51.0	3672	0.4372	0.4220	0.6169	0.4444	0.1702	0.2250	0.3888	0.6387	0.8152
0.3287	52.0	3744	0.4419	0.4246	0.6195	0.4488	0.1715	0.2276	0.3892	0.6375	0.8106
0.3307	53.0	3816	0.4567	0.4313	0.6200	0.4939	0.1755	0.2343	0.3788	0.6197	0.7940
0.3306	54.0	3888	0.4398	0.4239	0.6154	0.4522	0.1713	0.2264	0.3858	0.6316	0.8105
0.3549	55.0	3960	0.4450	0.4247	0.6155	0.4684	0.1721	0.2290	0.3822	0.6339	0.8157
0.3276	56.0	4032	0.4614	0.4458	0.6231	0.5053	0.1831	0.2360	0.3373	0.5794	0.7845
0.33	57.0	4104	0.4389	0.4244	0.6147	0.4517	0.1716	0.2259	0.3770	0.6359	0.8083
0.3338	58.0	4176	0.4326	0.4192	0.6174	0.4237	0.1684	0.2230	0.3956	0.6499	0.8116
0.3355	59.0	4248	0.4562	0.4436	0.6230	0.4861	0.1817	0.2331	0.3482	0.5884	0.7709
0.3173	60.0	4320	0.4317	0.4198	0.6143	0.4273	0.1688	0.2221	0.3898	0.6488	0.8106
0.3327	61.0	4392	0.4485	0.4341	0.6179	0.4691	0.1767	0.2291	0.3655	0.6056	0.7960
0.3132	62.0	4464	0.4433	0.4222	0.6144	0.4595	0.1707	0.2273	0.3852	0.6423	0.8137
0.3255	63.0	4536	0.4369	0.4216	0.6125	0.4406	0.1704	0.2235	0.3803	0.6408	0.8137
0.3264	64.0	4608	0.4292	0.4196	0.6139	0.4209	0.1680	0.2199	0.3936	0.6452	0.8133
0.3212	65.0	4680	0.4347	0.4224	0.6140	0.4383	0.1708	0.2235	0.3753	0.6398	0.8163
0.3102	66.0	4752	0.4373	0.4233	0.6138	0.4485	0.1710	0.2252	0.3818	0.6409	0.8054
0.337	67.0	4824	0.4400	0.4261	0.6149	0.4482	0.1726	0.2248	0.3725	0.6268	0.8113
0.3227	68.0	4896	0.4422	0.4241	0.6137	0.4573	0.1714	0.2263	0.3828	0.6378	0.8013
0.3537	69.0	4968	0.4310	0.4168	0.6112	0.4299	0.1677	0.2215	0.3919	0.6547	0.8127
0.3198	70.0	5040	0.4277	0.4161	0.6105	0.4206	0.1668	0.2194	0.3916	0.6578	0.8149
0.3353	71.0	5112	0.4375	0.4220	0.6128	0.4450	0.1702	0.2242	0.3813	0.6422	0.8137
0.3327	72.0	5184	0.4348	0.4210	0.6119	0.4401	0.1700	0.2229	0.3803	0.6421	0.8124
0.3145	73.0	5256	0.4354	0.4250	0.6136	0.4385	0.1717	0.2230	0.3704	0.6323	0.8099
0.3297	74.0	5328	0.4495	0.4316	0.6167	0.4775	0.1762	0.2304	0.3600	0.6149	0.8091
0.3192	75.0	5400	0.4440	0.4268	0.6149	0.4593	0.1731	0.2269	0.3763	0.6275	0.8057
0.3088	76.0	5472	0.4388	0.4236	0.6135	0.4440	0.1712	0.2241	0.3806	0.6337	0.8086
0.3311	77.0	5544	0.4503	0.4334	0.6181	0.4753	0.1767	0.2303	0.3636	0.6078	0.7981
0.3189	78.0	5616	0.4422	0.4273	0.6158	0.4566	0.1731	0.2267	0.3783	0.6246	0.8038
0.3365	79.0	5688	0.4453	0.4289	0.6160	0.4723	0.1747	0.2293	0.3652	0.6212	0.8129
0.3263	80.0	5760	0.4352	0.4230	0.6135	0.4412	0.1708	0.2235	0.3818	0.6327	0.8087
0.3354	81.0	5832	0.4308	0.4200	0.6130	0.4296	0.1689	0.2214	0.3876	0.6447	0.8130
0.3317	82.0	5904	0.4495	0.4353	0.6186	0.4743	0.1777	0.2301	0.3558	0.6035	0.8017
0.3074	83.0	5976	0.4472	0.4338	0.6179	0.4678	0.1764	0.2288	0.3599	0.6102	0.7947
0.3166	84.0	6048	0.4382	0.4249	0.6148	0.4459	0.1719	0.2246	0.3773	0.6288	0.8086
0.3318	85.0	6120	0.4362	0.4246	0.6148	0.4402	0.1713	0.2236	0.3815	0.6315	0.8071
0.3126	86.0	6192	0.4375	0.4244	0.6148	0.4480	0.1717	0.2249	0.3781	0.6316	0.8152
0.311	87.0	6264	0.4340	0.4216	0.6130	0.4398	0.1703	0.2232	0.3792	0.6393	0.8155
0.3092	88.0	6336	0.4263	0.4149	0.6109	0.4207	0.1663	0.2195	0.3988	0.6581	0.8160
0.2993	89.0	6408	0.4386	0.4276	0.6149	0.4493	0.1733	0.2247	0.3647	0.6224	0.8109
0.3234	90.0	6480	0.4333	0.4220	0.6132	0.4375	0.1704	0.2228	0.3814	0.6361	0.8132
0.3231	91.0	6552	0.4337	0.4237	0.6139	0.4387	0.1713	0.2230	0.3758	0.6337	0.8115
0.3097	92.0	6624	0.4374	0.4286	0.6153	0.4486	0.1740	0.2250	0.3644	0.6186	0.8102
0.3205	93.0	6696	0.4288	0.4197	0.6118	0.4278	0.1691	0.2208	0.3850	0.6438	0.8136
0.3098	94.0	6768	0.4361	0.4273	0.6162	0.4443	0.1729	0.2242	0.3682	0.6259	0.8096
0.3044	95.0	6840	0.4368	0.4271	0.6145	0.4459	0.1730	0.2244	0.3671	0.6255	0.8117
0.3205	96.0	6912	0.4326	0.4232	0.6134	0.4363	0.1710	0.2226	0.3763	0.6348	0.8127
0.2963	97.0	6984	0.4352	0.4257	0.6142	0.4413	0.1723	0.2235	0.3718	0.6273	0.8112
0.3048	98.0	7056	0.4350	0.4254	0.6142	0.4422	0.1722	0.2237	0.3710	0.6286	0.8128
0.3295	99.0	7128	0.4321	0.4225	0.6128	0.4349	0.1706	0.2222	0.3794	0.6351	0.8130
0.3093	100.0	7200	0.4346	0.4251	0.6137	0.4412	0.1720	0.2234	0.3728	0.6287	0.8112

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-230124-035129

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js