dropoff-utcustom-train-SF-RGB-b0_7

sam1120/dropoff-utcustom-train-SF-RGB-b0_7

dropoff-utcustom-train-SF-RGB-b0_7 is a image segmentation model from sam1120. Use it for the image segmentation task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as other.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

1% of all-time downloads

All-time downloads

525

Public

Repo size

419 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt119 MB · 61%

At a glance

Task: Image Segmentation
Library: transformers
License: other
Model type: segformer
Access: Public
Created: Feb 12, 2024
Updated: Feb 12, 2024
SHA: e00e10a7

Model card

From the Hugging Face model README

dropoff-utcustom-train-SF-RGB-b0_7

This model is a fine-tuned version of nvidia/mit-b0 on the sam1120/dropoff-utcustom-TRAIN dataset. It achieves the following results on the evaluation set:

Loss: 0.1457
Mean Iou: 0.6795
Mean Accuracy: 0.7207
Overall Accuracy: 0.9691
Accuracy Unlabeled: nan
Accuracy Dropoff: 0.4481
Accuracy Undropoff: 0.9932
Iou Unlabeled: nan
Iou Dropoff: 0.3907
Iou Undropoff: 0.9684

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: 9e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 120

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Dropoff	Accuracy Undropoff	Iou Unlabeled	Iou Dropoff	Iou Undropoff
1.1505	3.33	10	1.1103	0.1106	0.6036	0.2919	nan	0.9456	0.2616	0.0	0.0703	0.2616
0.9635	6.67	20	1.0114	0.3710	0.8470	0.8737	nan	0.8177	0.8763	0.0	0.2435	0.8694
0.9358	10.0	30	0.8242	0.4206	0.7727	0.9440	nan	0.5848	0.9606	0.0	0.3194	0.9425
0.579	13.33	40	0.5703	0.4525	0.7615	0.9633	nan	0.5402	0.9829	0.0	0.3951	0.9624
0.4411	16.67	50	0.4166	0.4529	0.7380	0.9667	nan	0.4872	0.9889	0.0	0.3928	0.9659
0.4311	20.0	60	0.3843	0.6678	0.7156	0.9667	nan	0.4400	0.9911	nan	0.3695	0.9661
0.3437	23.33	70	0.3590	0.4347	0.6956	0.9655	nan	0.3995	0.9918	0.0	0.3392	0.9649
0.3136	26.67	80	0.3198	0.6259	0.6622	0.9638	nan	0.3312	0.9931	nan	0.2885	0.9633
0.2682	30.0	90	0.2919	0.6187	0.6470	0.9648	nan	0.2984	0.9957	nan	0.2730	0.9643
0.2521	33.33	100	0.2957	0.6448	0.6845	0.9653	nan	0.3764	0.9926	nan	0.3248	0.9648
0.2287	36.67	110	0.2747	0.6800	0.7256	0.9685	nan	0.4591	0.9921	nan	0.3922	0.9678
0.2203	40.0	120	0.2537	0.7108	0.7687	0.9706	nan	0.5472	0.9902	nan	0.4517	0.9699
0.1964	43.33	130	0.2356	0.6689	0.7054	0.9686	nan	0.4167	0.9941	nan	0.3699	0.9680
0.1776	46.67	140	0.2205	0.6729	0.7137	0.9684	nan	0.4343	0.9931	nan	0.3780	0.9677
0.1675	50.0	150	0.2061	0.6809	0.7244	0.9689	nan	0.4562	0.9926	nan	0.3936	0.9682
0.148	53.33	160	0.1954	0.6924	0.7418	0.9694	nan	0.4920	0.9915	nan	0.4160	0.9687
0.1364	56.67	170	0.1915	0.6869	0.7415	0.9681	nan	0.4928	0.9902	nan	0.4064	0.9674
0.1171	60.0	180	0.1776	0.7206	0.7816	0.9714	nan	0.5734	0.9899	nan	0.4706	0.9707
0.1169	63.33	190	0.1754	0.6580	0.6853	0.9689	nan	0.3741	0.9965	nan	0.3476	0.9684
0.1178	66.67	200	0.1676	0.6783	0.7233	0.9684	nan	0.4545	0.9922	nan	0.3888	0.9677
0.1016	70.0	210	0.1670	0.6633	0.6985	0.9682	nan	0.4025	0.9944	nan	0.3590	0.9676
0.1025	73.33	220	0.1648	0.6789	0.7154	0.9696	nan	0.4366	0.9943	nan	0.3888	0.9690
0.0956	76.67	230	0.1607	0.6684	0.7103	0.9677	nan	0.4279	0.9927	nan	0.3697	0.9671
0.1443	80.0	240	0.1611	0.6747	0.7134	0.9688	nan	0.4332	0.9937	nan	0.3811	0.9682
0.0902	83.33	250	0.1600	0.6713	0.7060	0.9691	nan	0.4174	0.9946	nan	0.3740	0.9685
0.0846	86.67	260	0.1559	0.6772	0.7263	0.9677	nan	0.4613	0.9912	nan	0.3874	0.9670
0.1166	90.0	270	0.1587	0.6615	0.6984	0.9677	nan	0.4030	0.9939	nan	0.3559	0.9671
0.0825	93.33	280	0.1538	0.6684	0.7068	0.9682	nan	0.4199	0.9936	nan	0.3692	0.9676
0.0769	96.67	290	0.1527	0.6649	0.7033	0.9679	nan	0.4130	0.9936	nan	0.3626	0.9673
0.0722	100.0	300	0.1473	0.6832	0.7247	0.9694	nan	0.4563	0.9932	nan	0.3976	0.9688
0.0779	103.33	310	0.1465	0.6809	0.7200	0.9695	nan	0.4462	0.9937	nan	0.3930	0.9689
0.0771	106.67	320	0.1494	0.6673	0.7052	0.9682	nan	0.4167	0.9937	nan	0.3670	0.9676
0.1082	110.0	330	0.1479	0.6753	0.7182	0.9683	nan	0.4438	0.9926	nan	0.3830	0.9677
0.0726	113.33	340	0.1451	0.6765	0.7159	0.9689	nan	0.4384	0.9935	nan	0.3846	0.9683
0.0743	116.67	350	0.1469	0.6814	0.7249	0.9689	nan	0.4571	0.9927	nan	0.3946	0.9683
0.0703	120.0	360	0.1457	0.6795	0.7207	0.9691	nan	0.4481	0.9932	nan	0.3907	0.9684

Framework versions

Transformers 4.30.2
Pytorch 2.0.1+cu117
Datasets 2.13.1
Tokenizers 0.13.3

Task: Image Segmentation
Library: transformers
Type: segformer
License: other
Created: Feb 12, 2024
Updated: Feb 12, 2024