glpn-nyu-finetuned-diode-230125-042306

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: 0.4380
Mae: 0.4255
Rmse: 0.6150
Abs Rel: 0.4444
Log Mae: 0.1724
Log Rmse: 0.2247
Delta1: 0.3675
Delta2: 0.6329
Delta3: 0.8147

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	0.4779	0.6689	0.5504	0.2005	0.2590	0.3023	0.5336	0.8000
0.4776	2.0	144	0.4638	0.4495	0.6305	0.4854	0.1854	0.2371	0.3323	0.5842	0.7749
0.4668	3.0	216	0.4843	0.4705	0.6368	0.5459	0.1961	0.2469	0.3115	0.5258	0.7237
0.439	4.0	288	0.4596	0.4383	0.6224	0.4903	0.1794	0.2347	0.3564	0.6054	0.7900
0.4629	5.0	360	0.4846	0.4622	0.6347	0.5505	0.1914	0.2466	0.3240	0.5567	0.7432
0.4557	6.0	432	0.4660	0.4399	0.6223	0.5107	0.1801	0.2373	0.3605	0.5922	0.7992
0.4131	7.0	504	0.4737	0.4466	0.6291	0.4877	0.1847	0.2387	0.3592	0.5753	0.7545
0.3742	8.0	576	0.4756	0.4555	0.6363	0.5127	0.1879	0.2424	0.3462	0.5642	0.7581
0.3943	9.0	648	0.4816	0.4606	0.6340	0.5566	0.1901	0.2459	0.3304	0.5512	0.7484
0.3699	10.0	720	0.4779	0.4527	0.6289	0.5402	0.1869	0.2433	0.3419	0.5659	0.7699
0.3695	11.0	792	0.4335	0.4185	0.6141	0.4174	0.1685	0.2210	0.3837	0.6484	0.8142
0.4268	12.0	864	0.4831	0.4622	0.6345	0.5491	0.1912	0.2456	0.3283	0.5515	0.7369
0.4295	13.0	936	0.4512	0.4421	0.6267	0.4498	0.1803	0.2292	0.3508	0.5951	0.7803
0.4071	14.0	1008	0.4632	0.4514	0.6295	0.4755	0.1842	0.2334	0.3514	0.5676	0.7346
0.4383	15.0	1080	0.4655	0.4394	0.6283	0.4894	0.1793	0.2370	0.3762	0.6022	0.7816
0.4009	16.0	1152	0.4684	0.4434	0.6294	0.5215	0.1814	0.2403	0.3601	0.5881	0.7980
0.3889	17.0	1224	0.4619	0.4379	0.6357	0.4623	0.1791	0.2389	0.3946	0.6088	0.7665
0.4114	18.0	1296	0.4838	0.4642	0.6358	0.5514	0.1924	0.2471	0.3310	0.5444	0.7336
0.3656	19.0	1368	0.4771	0.4524	0.6317	0.5284	0.1869	0.2428	0.3379	0.5765	0.7665
0.4117	20.0	1440	0.4388	0.4187	0.6257	0.4113	0.1680	0.2270	0.4162	0.6619	0.8001
0.3641	21.0	1512	0.4593	0.4374	0.6238	0.4773	0.1779	0.2332	0.3705	0.6088	0.7745
0.3559	22.0	1584	0.4534	0.4300	0.6242	0.4663	0.1747	0.2329	0.3854	0.6288	0.7987
0.3897	23.0	1656	0.4695	0.4506	0.6292	0.5215	0.1852	0.2404	0.3432	0.5746	0.7698
0.4281	24.0	1728	0.4920	0.4693	0.6380	0.5835	0.1949	0.2514	0.3239	0.5352	0.7230
0.4113	25.0	1800	0.4525	0.4335	0.6405	0.4109	0.1757	0.2330	0.4046	0.6251	0.7878
0.3734	26.0	1872	0.4357	0.4159	0.6203	0.4158	0.1667	0.2241	0.4234	0.6609	0.7919
0.3408	27.0	1944	0.4544	0.4419	0.6257	0.4712	0.1806	0.2325	0.3525	0.5993	0.7850
0.3816	28.0	2016	0.4622	0.4465	0.6252	0.4919	0.1823	0.2346	0.3465	0.5844	0.7687
0.3643	29.0	2088	0.4534	0.4370	0.6219	0.4721	0.1778	0.2311	0.3653	0.6016	0.7886
0.3762	30.0	2160	0.4418	0.4302	0.6209	0.4394	0.1745	0.2261	0.3724	0.6226	0.7944
0.3704	31.0	2232	0.4723	0.4496	0.6271	0.5262	0.1848	0.2406	0.3477	0.5726	0.7679
0.3657	32.0	2304	0.4458	0.4311	0.6188	0.4580	0.1755	0.2283	0.3641	0.6167	0.8132
0.4261	33.0	2376	0.4551	0.4360	0.6240	0.4757	0.1778	0.2333	0.3707	0.6109	0.7859
0.3499	34.0	2448	0.4297	0.4131	0.6154	0.4141	0.1654	0.2222	0.4208	0.6585	0.8011
0.3316	35.0	2520	0.4553	0.4368	0.6200	0.4786	0.1780	0.2317	0.3625	0.6038	0.7848
0.3468	36.0	2592	0.4430	0.4275	0.6159	0.4460	0.1732	0.2253	0.3776	0.6204	0.8069
0.3439	37.0	2664	0.4550	0.4353	0.6234	0.4678	0.1772	0.2319	0.3741	0.6089	0.7857
0.3854	38.0	2736	0.4619	0.4410	0.6238	0.4960	0.1806	0.2359	0.3556	0.5983	0.7832
0.3521	39.0	2808	0.4743	0.4607	0.6317	0.5248	0.1902	0.2412	0.3241	0.5544	0.7351
0.3836	40.0	2880	0.4701	0.4508	0.6264	0.5249	0.1856	0.2399	0.3364	0.5747	0.7680
0.3601	41.0	2952	0.4749	0.4551	0.6281	0.5289	0.1879	0.2412	0.3288	0.5613	0.7672
0.3552	42.0	3024	0.4403	0.4215	0.6224	0.4299	0.1697	0.2267	0.4062	0.6517	0.8015
0.3582	43.0	3096	0.4307	0.4170	0.6174	0.4187	0.1676	0.2229	0.4009	0.6648	0.8095
0.332	44.0	3168	0.4663	0.4462	0.6244	0.5113	0.1834	0.2376	0.3452	0.5794	0.7755
0.3407	45.0	3240	0.4491	0.4333	0.6202	0.4714	0.1770	0.2309	0.3514	0.6155	0.8089
0.3613	46.0	3312	0.4767	0.4539	0.6282	0.5360	0.1874	0.2423	0.3333	0.5698	0.7528
0.3729	47.0	3384	0.4647	0.4435	0.6244	0.5128	0.1822	0.2381	0.3471	0.5923	0.7886
0.3304	48.0	3456	0.4431	0.4285	0.6150	0.4599	0.1739	0.2266	0.3627	0.6212	0.8095
0.357	49.0	3528	0.4558	0.4372	0.6219	0.4788	0.1784	0.2324	0.3579	0.6054	0.7861
0.3548	50.0	3600	0.4482	0.4308	0.6197	0.4612	0.1753	0.2295	0.3663	0.6237	0.8060
0.3332	51.0	3672	0.4533	0.4317	0.6252	0.4710	0.1755	0.2330	0.3745	0.6278	0.7971
0.3369	52.0	3744	0.4350	0.4189	0.6203	0.4229	0.1683	0.2249	0.4017	0.6581	0.8048
0.3379	53.0	3816	0.4344	0.4192	0.6192	0.4275	0.1683	0.2242	0.3953	0.6563	0.8049
0.3237	54.0	3888	0.4554	0.4392	0.6223	0.4822	0.1798	0.2328	0.3529	0.5952	0.7919
0.3523	55.0	3960	0.4511	0.4350	0.6207	0.4752	0.1771	0.2311	0.3673	0.6043	0.7962
0.326	56.0	4032	0.4460	0.4327	0.6208	0.4581	0.1756	0.2282	0.3644	0.6160	0.8041
0.3214	57.0	4104	0.4397	0.4252	0.6160	0.4384	0.1717	0.2241	0.3749	0.6333	0.8019
0.3342	58.0	4176	0.4493	0.4316	0.6176	0.4685	0.1754	0.2291	0.3640	0.6201	0.7951
0.3361	59.0	4248	0.4568	0.4394	0.6215	0.4935	0.1798	0.2341	0.3509	0.5953	0.7997
0.3141	60.0	4320	0.4425	0.4270	0.6182	0.4459	0.1727	0.2265	0.3829	0.6222	0.7972
0.3395	61.0	4392	0.4397	0.4229	0.6138	0.4450	0.1707	0.2246	0.3807	0.6318	0.8108
0.3124	62.0	4464	0.4232	0.4104	0.6128	0.4073	0.1641	0.2192	0.4074	0.6707	0.8209
0.3106	63.0	4536	0.4426	0.4223	0.6156	0.4504	0.1708	0.2267	0.3869	0.6404	0.8063
0.3268	64.0	4608	0.4391	0.4242	0.6160	0.4409	0.1715	0.2248	0.3818	0.6346	0.8082
0.3153	65.0	4680	0.4558	0.4355	0.6204	0.4877	0.1779	0.2333	0.3607	0.6069	0.8013
0.3063	66.0	4752	0.4367	0.4206	0.6154	0.4402	0.1694	0.2246	0.3891	0.6475	0.8129
0.3327	67.0	4824	0.4668	0.4466	0.6246	0.5172	0.1834	0.2383	0.3465	0.5778	0.7821
0.3189	68.0	4896	0.4423	0.4265	0.6171	0.4531	0.1726	0.2267	0.3748	0.6261	0.8109
0.3241	69.0	4968	0.4606	0.4433	0.6227	0.5013	0.1817	0.2353	0.3480	0.5843	0.7906
0.3165	70.0	5040	0.4359	0.4222	0.6128	0.4366	0.1702	0.2229	0.3809	0.6371	0.8136
0.3293	71.0	5112	0.4289	0.4150	0.6109	0.4181	0.1666	0.2197	0.3948	0.6586	0.8183
0.3256	72.0	5184	0.4457	0.4295	0.6174	0.4632	0.1747	0.2286	0.3657	0.6209	0.8117
0.3129	73.0	5256	0.4481	0.4314	0.6178	0.4680	0.1755	0.2291	0.3597	0.6201	0.8060
0.3197	74.0	5328	0.4365	0.4228	0.6150	0.4400	0.1706	0.2240	0.3744	0.6410	0.8159
0.323	75.0	5400	0.4351	0.4221	0.6137	0.4352	0.1703	0.2230	0.3752	0.6392	0.8141
0.3087	76.0	5472	0.4342	0.4215	0.6155	0.4321	0.1701	0.2232	0.3765	0.6439	0.8180
0.3126	77.0	5544	0.4362	0.4247	0.6160	0.4377	0.1717	0.2241	0.3731	0.6397	0.8094
0.3185	78.0	5616	0.4377	0.4234	0.6163	0.4446	0.1713	0.2256	0.3737	0.6433	0.8163
0.3195	79.0	5688	0.4426	0.4265	0.6174	0.4576	0.1731	0.2280	0.3734	0.6336	0.8149
0.3173	80.0	5760	0.4415	0.4259	0.6168	0.4550	0.1725	0.2273	0.3714	0.6381	0.8135
0.3207	81.0	5832	0.4374	0.4258	0.6172	0.4402	0.1722	0.2249	0.3689	0.6359	0.8095
0.3258	82.0	5904	0.4405	0.4283	0.6173	0.4445	0.1737	0.2257	0.3646	0.6299	0.8078
0.2971	83.0	5976	0.4430	0.4307	0.6185	0.4529	0.1748	0.2271	0.3604	0.6259	0.8040
0.3132	84.0	6048	0.4423	0.4277	0.6157	0.4478	0.1732	0.2252	0.3703	0.6270	0.8049
0.3281	85.0	6120	0.4378	0.4240	0.6152	0.4368	0.1713	0.2238	0.3770	0.6407	0.8108
0.3023	86.0	6192	0.4371	0.4241	0.6145	0.4405	0.1715	0.2241	0.3726	0.6370	0.8153
0.3051	87.0	6264	0.4327	0.4194	0.6136	0.4288	0.1692	0.2222	0.3798	0.6511	0.8170
0.3076	88.0	6336	0.4319	0.4175	0.6122	0.4262	0.1680	0.2215	0.3889	0.6534	0.8183
0.2981	89.0	6408	0.4374	0.4244	0.6136	0.4402	0.1716	0.2236	0.3728	0.6331	0.8140
0.3238	90.0	6480	0.4349	0.4222	0.6136	0.4371	0.1706	0.2233	0.3743	0.6418	0.8195
0.32	91.0	6552	0.4375	0.4240	0.6143	0.4417	0.1715	0.2242	0.3717	0.6379	0.8165
0.3087	92.0	6624	0.4421	0.4288	0.6162	0.4531	0.1739	0.2263	0.3652	0.6244	0.8125
0.3207	93.0	6696	0.4352	0.4216	0.6129	0.4376	0.1702	0.2231	0.3782	0.6406	0.8184
0.3064	94.0	6768	0.4398	0.4259	0.6148	0.4478	0.1727	0.2252	0.3685	0.6300	0.8147
0.3076	95.0	6840	0.4385	0.4258	0.6147	0.4446	0.1724	0.2246	0.3669	0.6321	0.8135
0.3181	96.0	6912	0.4393	0.4262	0.6150	0.4471	0.1728	0.2251	0.3663	0.6306	0.8147
0.2956	97.0	6984	0.4392	0.4271	0.6156	0.4470	0.1731	0.2252	0.3650	0.6297	0.8141
0.3026	98.0	7056	0.4390	0.4260	0.6151	0.4462	0.1726	0.2250	0.3669	0.6317	0.8144
0.329	99.0	7128	0.4362	0.4242	0.6156	0.4389	0.1713	0.2238	0.3716	0.6380	0.8156
0.3095	100.0	7200	0.4380	0.4255	0.6150	0.4444	0.1724	0.2247	0.3675	0.6329	0.8147

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-230125-042306

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js