glpn-nyu-finetuned-diode-230103-091356

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: 0.4360
Mae: 0.4251
Rmse: 0.6169
Abs Rel: 0.4500
Log Mae: 0.1721
Log Rmse: 0.2269
Delta1: 0.3828
Delta2: 0.6326
Delta3: 0.8051

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
1.0762	1.0	72	0.5031	0.4779	0.6690	0.5503	0.2006	0.2591	0.3020	0.5337	0.8000
0.478	2.0	144	0.4653	0.4509	0.6307	0.4891	0.1861	0.2377	0.3300	0.5805	0.7734
0.4668	3.0	216	0.4845	0.4712	0.6373	0.5469	0.1963	0.2471	0.3110	0.5254	0.7235
0.4389	4.0	288	0.4587	0.4368	0.6219	0.4887	0.1787	0.2344	0.3578	0.6099	0.7926
0.4626	5.0	360	0.4879	0.4662	0.6351	0.5617	0.1937	0.2482	0.3135	0.5462	0.7395
0.4534	6.0	432	0.4638	0.4422	0.6236	0.4951	0.1810	0.2358	0.3606	0.5844	0.7831
0.4108	7.0	504	0.4688	0.4508	0.6279	0.5050	0.1856	0.2385	0.3426	0.5701	0.7623
0.3832	8.0	576	0.4759	0.4533	0.6284	0.5257	0.1869	0.2411	0.3331	0.5701	0.7617
0.4097	9.0	648	0.4771	0.4501	0.6303	0.5361	0.1855	0.2433	0.3454	0.5838	0.7609
0.3799	10.0	720	0.4575	0.4375	0.6240	0.4874	0.1790	0.2349	0.3669	0.6032	0.7916
0.3659	11.0	792	0.4718	0.4590	0.6298	0.5176	0.1893	0.2396	0.3283	0.5502	0.7368
0.4145	12.0	864	0.4776	0.4561	0.6298	0.5325	0.1883	0.2421	0.3333	0.5611	0.7540
0.4224	13.0	936	0.4320	0.4138	0.6202	0.4013	0.1655	0.2232	0.4217	0.6641	0.8004
0.4142	14.0	1008	0.4597	0.4440	0.6234	0.4842	0.1813	0.2330	0.3520	0.5895	0.7617
0.4393	15.0	1080	0.4333	0.4251	0.6197	0.4182	0.1712	0.2225	0.3787	0.6303	0.8100
0.4045	16.0	1152	0.4603	0.4356	0.6197	0.4819	0.1776	0.2322	0.3635	0.6050	0.7858
0.3708	17.0	1224	0.4738	0.4567	0.6292	0.5264	0.1886	0.2411	0.3283	0.5557	0.7596
0.4042	18.0	1296	0.5004	0.4802	0.6423	0.6101	0.2008	0.2560	0.3022	0.5165	0.6931
0.3763	19.0	1368	0.4501	0.4361	0.6213	0.4723	0.1772	0.2303	0.3634	0.6034	0.7889
0.4084	20.0	1440	0.4272	0.4133	0.6208	0.3958	0.1649	0.2226	0.4284	0.6684	0.8009
0.3637	21.0	1512	0.4307	0.4145	0.6199	0.4134	0.1665	0.2241	0.3957	0.6847	0.8137
0.3655	22.0	1584	0.4591	0.4374	0.6370	0.4594	0.1791	0.2384	0.3816	0.6264	0.7826
0.3844	23.0	1656	0.4692	0.4444	0.6273	0.5241	0.1824	0.2407	0.3540	0.5990	0.7756
0.428	24.0	1728	0.4982	0.4753	0.6403	0.6084	0.1984	0.2552	0.3099	0.5233	0.7204
0.4051	25.0	1800	0.4824	0.4618	0.6329	0.5533	0.1915	0.2461	0.3248	0.5495	0.7415
0.3584	26.0	1872	0.4434	0.4207	0.6177	0.4468	0.1694	0.2277	0.3975	0.6442	0.8038
0.3443	27.0	1944	0.4602	0.4434	0.6241	0.4912	0.1822	0.2351	0.3431	0.5877	0.7893
0.3714	28.0	2016	0.4818	0.4594	0.6316	0.5521	0.1900	0.2455	0.3283	0.5567	0.7493
0.3688	29.0	2088	0.4443	0.4215	0.6242	0.4386	0.1702	0.2294	0.4024	0.6522	0.8065
0.3615	30.0	2160	0.4462	0.4291	0.6189	0.4500	0.1739	0.2277	0.3792	0.6208	0.7896
0.3655	31.0	2232	0.4808	0.4574	0.6305	0.5524	0.1893	0.2452	0.3322	0.5590	0.7460
0.3576	32.0	2304	0.4321	0.4102	0.6182	0.4079	0.1640	0.2241	0.4296	0.6713	0.8074
0.3947	33.0	2376	0.4468	0.4298	0.6232	0.4574	0.1744	0.2306	0.3873	0.6163	0.7873
0.3402	34.0	2448	0.4565	0.4352	0.6195	0.4913	0.1776	0.2337	0.3734	0.6039	0.7865
0.3412	35.0	2520	0.4438	0.4261	0.6180	0.4546	0.1728	0.2279	0.3778	0.6252	0.8043
0.3547	36.0	2592	0.4577	0.4416	0.6218	0.4868	0.1807	0.2329	0.3517	0.5862	0.7862
0.3425	37.0	2664	0.4682	0.4511	0.6285	0.5210	0.1860	0.2406	0.3411	0.5748	0.7694
0.3853	38.0	2736	0.4752	0.4514	0.6289	0.5458	0.1863	0.2438	0.3408	0.5721	0.7760
0.3643	39.0	2808	0.4737	0.4547	0.6291	0.5401	0.1875	0.2428	0.3316	0.5673	0.7617
0.398	40.0	2880	0.4662	0.4467	0.6274	0.5124	0.1838	0.2394	0.3514	0.5823	0.7700
0.3579	41.0	2952	0.4781	0.4545	0.6290	0.5513	0.1880	0.2446	0.3343	0.5624	0.7718
0.3545	42.0	3024	0.4460	0.4277	0.6221	0.4553	0.1730	0.2294	0.3862	0.6285	0.7999
0.3527	43.0	3096	0.4330	0.4153	0.6169	0.4221	0.1668	0.2240	0.4106	0.6618	0.8084
0.3251	44.0	3168	0.4503	0.4286	0.6172	0.4781	0.1744	0.2313	0.3725	0.6224	0.8095
0.3433	45.0	3240	0.4471	0.4346	0.6187	0.4652	0.1772	0.2293	0.3606	0.6043	0.7952
0.3607	46.0	3312	0.4474	0.4263	0.6166	0.4658	0.1728	0.2293	0.3835	0.6287	0.8039
0.3722	47.0	3384	0.4527	0.4337	0.6205	0.4857	0.1768	0.2329	0.3696	0.6084	0.7922
0.3322	48.0	3456	0.4629	0.4431	0.6236	0.5118	0.1818	0.2373	0.3460	0.5897	0.7954
0.3624	49.0	3528	0.4431	0.4304	0.6203	0.4511	0.1742	0.2277	0.3827	0.6152	0.7917
0.3386	50.0	3600	0.4475	0.4260	0.6173	0.4697	0.1727	0.2301	0.3870	0.6283	0.8102
0.3316	51.0	3672	0.4558	0.4328	0.6194	0.4982	0.1770	0.2345	0.3618	0.6120	0.8124
0.3259	52.0	3744	0.4316	0.4084	0.6165	0.4234	0.1630	0.2245	0.4311	0.6809	0.8148
0.3299	53.0	3816	0.4489	0.4222	0.6198	0.4779	0.1706	0.2327	0.4049	0.6441	0.8021
0.3334	54.0	3888	0.4831	0.4598	0.6319	0.5716	0.1902	0.2476	0.3281	0.5597	0.7549
0.3342	55.0	3960	0.4478	0.4288	0.6166	0.4786	0.1745	0.2310	0.3749	0.6218	0.8091
0.3276	56.0	4032	0.4524	0.4342	0.6192	0.4852	0.1773	0.2326	0.3596	0.6113	0.8007
0.326	57.0	4104	0.4411	0.4226	0.6162	0.4486	0.1704	0.2268	0.3947	0.6403	0.7959
0.3429	58.0	4176	0.4578	0.4418	0.6221	0.4961	0.1812	0.2349	0.3497	0.5956	0.7750
0.3347	59.0	4248	0.4586	0.4409	0.6220	0.4946	0.1808	0.2347	0.3439	0.6004	0.7869
0.3215	60.0	4320	0.4583	0.4382	0.6232	0.4974	0.1789	0.2357	0.3667	0.6008	0.7855
0.331	61.0	4392	0.4412	0.4206	0.6145	0.4579	0.1699	0.2276	0.3966	0.6413	0.8047
0.3124	62.0	4464	0.4455	0.4236	0.6181	0.4727	0.1715	0.2313	0.3902	0.6417	0.8098
0.322	63.0	4536	0.4406	0.4230	0.6143	0.4548	0.1716	0.2269	0.3775	0.6425	0.8115
0.3194	64.0	4608	0.4473	0.4331	0.6193	0.4657	0.1765	0.2297	0.3606	0.6122	0.8014
0.3159	65.0	4680	0.4407	0.4225	0.6186	0.4548	0.1712	0.2293	0.3913	0.6433	0.8075
0.3118	66.0	4752	0.4478	0.4258	0.6169	0.4801	0.1728	0.2315	0.3762	0.6391	0.8064
0.336	67.0	4824	0.4659	0.4463	0.6252	0.5210	0.1834	0.2394	0.3464	0.5820	0.7786
0.3233	68.0	4896	0.4370	0.4208	0.6168	0.4452	0.1696	0.2265	0.4019	0.6425	0.8059
0.3285	69.0	4968	0.4479	0.4340	0.6189	0.4773	0.1771	0.2312	0.3609	0.6136	0.7972
0.3186	70.0	5040	0.4469	0.4308	0.6198	0.4698	0.1751	0.2310	0.3741	0.6219	0.7966
0.3351	71.0	5112	0.4476	0.4292	0.6176	0.4769	0.1745	0.2311	0.3718	0.6220	0.8035
0.3286	72.0	5184	0.4415	0.4229	0.6155	0.4655	0.1713	0.2289	0.3816	0.6376	0.8117
0.3135	73.0	5256	0.4527	0.4335	0.6198	0.4918	0.1769	0.2338	0.3621	0.6152	0.8036
0.3244	74.0	5328	0.4449	0.4290	0.6171	0.4685	0.1746	0.2296	0.3667	0.6234	0.8073
0.3253	75.0	5400	0.4450	0.4303	0.6182	0.4680	0.1750	0.2296	0.3703	0.6185	0.8013
0.3072	76.0	5472	0.4312	0.4212	0.6161	0.4337	0.1700	0.2242	0.3840	0.6411	0.8104
0.3159	77.0	5544	0.4434	0.4314	0.6186	0.4636	0.1754	0.2290	0.3643	0.6171	0.7996
0.3176	78.0	5616	0.4319	0.4207	0.6177	0.4330	0.1695	0.2249	0.3889	0.6524	0.8080
0.3243	79.0	5688	0.4432	0.4304	0.6186	0.4698	0.1752	0.2302	0.3667	0.6218	0.8058
0.3183	80.0	5760	0.4438	0.4288	0.6175	0.4665	0.1742	0.2294	0.3730	0.6235	0.8030
0.323	81.0	5832	0.4365	0.4248	0.6170	0.4480	0.1716	0.2263	0.3820	0.6313	0.8056
0.3348	82.0	5904	0.4385	0.4280	0.6179	0.4532	0.1738	0.2273	0.3651	0.6249	0.8099
0.2948	83.0	5976	0.4456	0.4330	0.6190	0.4727	0.1763	0.2305	0.3622	0.6121	0.7981
0.3156	84.0	6048	0.4349	0.4236	0.6155	0.4442	0.1712	0.2252	0.3834	0.6331	0.8086
0.3227	85.0	6120	0.4352	0.4251	0.6160	0.4423	0.1719	0.2250	0.3799	0.6293	0.8055
0.3044	86.0	6192	0.4349	0.4235	0.6165	0.4444	0.1714	0.2259	0.3858	0.6312	0.8108
0.3067	87.0	6264	0.4293	0.4214	0.6150	0.4293	0.1700	0.2229	0.3862	0.6397	0.8102
0.3083	88.0	6336	0.4260	0.4164	0.6139	0.4229	0.1673	0.2221	0.3989	0.6536	0.8126
0.2989	89.0	6408	0.4381	0.4270	0.6168	0.4526	0.1731	0.2270	0.3766	0.6248	0.8051
0.3232	90.0	6480	0.4352	0.4230	0.6158	0.4480	0.1711	0.2263	0.3854	0.6358	0.8112
0.3201	91.0	6552	0.4361	0.4242	0.6164	0.4462	0.1718	0.2262	0.3842	0.6327	0.8078
0.3096	92.0	6624	0.4390	0.4273	0.6171	0.4563	0.1733	0.2279	0.3790	0.6237	0.8046
0.322	93.0	6696	0.4338	0.4229	0.6157	0.4447	0.1709	0.2258	0.3889	0.6351	0.8069
0.3096	94.0	6768	0.4348	0.4238	0.6160	0.4448	0.1714	0.2256	0.3839	0.6342	0.8077
0.3067	95.0	6840	0.4414	0.4298	0.6181	0.4628	0.1748	0.2290	0.3707	0.6205	0.8027
0.3198	96.0	6912	0.4334	0.4228	0.6162	0.4434	0.1709	0.2258	0.3872	0.6370	0.8077
0.295	97.0	6984	0.4367	0.4261	0.6169	0.4507	0.1728	0.2269	0.3791	0.6283	0.8045
0.305	98.0	7056	0.4373	0.4266	0.6171	0.4524	0.1730	0.2273	0.3781	0.6280	0.8046
0.3304	99.0	7128	0.4334	0.4230	0.6162	0.4432	0.1709	0.2257	0.3874	0.6378	0.8062
0.3099	100.0	7200	0.4360	0.4251	0.6169	0.4500	0.1721	0.2269	0.3828	0.6326	0.8051

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2

glpn-nyu-finetuned-diode-230103-091356

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js