segformer-b0-scene-parse-150-lr-4-e-30

klentree/segformer-b0-scene-parse-150-lr-4-e-30

segformer-b0-scene-parse-150-lr-4-e-30 is a machine learning model from klentree. Use it for the machine learning 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 openrail.

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

Downloads · 30 days

31% of all-time downloads

All-time downloads

Public

Parameters

47.2M

3.8 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors189 MB · 100%

At a glance

Library: transformers
License: openrail
Model type: segformer
Access: Public
Created: Jan 24, 2024
Updated: Jan 24, 2024
SHA: 2bd0a6f5

Base models

DiTo97/binarization-segformer-b3

Model card

From the Hugging Face model README

segformer-b0-scene-parse-150-lr-4-e-30

This model is a fine-tuned version of DiTo97/binarization-segformer-b3 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1880
Mean Iou: 0.5071
Mean Accuracy: 0.5450
Overall Accuracy: 0.9491
Per Category Iou: [0.06535303105902995, 0.9489053008277266]
Per Category Accuracy: [0.11405408708523163, 0.9759907496995939]

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: 0.0001
train_batch_size: 1
eval_batch_size: 1
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
No log	1.0	112	0.3289	0.4844	0.5000	0.9687	[4.9068792755415755e-05, 0.9686888107710686]	[4.92256299183194e-05, 0.9998970207890034]
No log	2.0	224	0.1407	0.4844	0.5	0.9688	[0.0, 0.9687870873345269]	[0.0, 1.0]
No log	3.0	336	0.1423	0.4846	0.5002	0.9688	[0.00048369030243372596, 0.9687966297488466]	[0.00048376912161107, 0.999994749864973]
No log	4.0	448	0.1608	0.4958	0.5118	0.9675	[0.024229354911501754, 0.9674411654710736]	[0.025880799288435032, 0.9978040216471817]
0.4177	5.0	560	0.1491	0.4954	0.5111	0.9682	[0.02251428850967379, 0.9682168520862131]	[0.023436492147663312, 0.998680301996517]
0.4177	6.0	672	0.1285	0.4964	0.5121	0.9684	[0.024459881085535998, 0.9683673031128758]	[0.025390240424766262, 0.9987745309824687]
0.4177	7.0	784	0.1574	0.5096	0.5317	0.9626	[0.0567450633090623, 0.9625025362567224]	[0.07210875808828023, 0.9912768459638821]
0.4177	8.0	896	0.1433	0.4987	0.5159	0.9656	[0.031883115624203075, 0.9655670514632985]	[0.03628947386288795, 0.9955472839192083]
0.1718	9.0	1008	0.1630	0.5049	0.5276	0.9604	[0.04946925848938199, 0.9603548723540993]	[0.06596743639709128, 0.9892549970894564]
0.1718	10.0	1120	0.1443	0.5096	0.5332	0.9612	[0.05814756470020897, 0.9611024416342511]	[0.0767529416557465, 0.9896910864301292]
0.1718	11.0	1232	0.1421	0.4987	0.5155	0.9663	[0.031053353658536585, 0.9662867772812311]	[0.034578458864347746, 0.9963426246868786]
0.1718	12.0	1344	0.1461	0.5021	0.5208	0.9642	[0.04000756200631968, 0.9641641173493553]	[0.04777601998900062, 0.9937439719152777]
0.1718	13.0	1456	0.1558	0.5139	0.5437	0.9578	[0.07017424997284097, 0.957692008002001]	[0.10197174109355586, 0.9854010698462651]
0.1428	14.0	1568	0.1611	0.5119	0.5467	0.9535	[0.07044097921924365, 0.9533510471833032]	[0.11285569761204772, 0.9806002041865014]
0.1428	15.0	1680	0.1797	0.5102	0.5619	0.9425	[0.07812492030758297, 0.9422603916355587]	[0.15600450838872632, 0.9678826271063159]
0.1428	16.0	1792	0.1670	0.5067	0.5414	0.9511	[0.06249891980313476, 0.9509835090222434]	[0.104349848249265, 0.9784256091961365]
0.1428	17.0	1904	0.1762	0.5080	0.5516	0.9459	[0.07031706424823504, 0.9456499586079702]	[0.13116253963511926, 0.9721212361180415]
0.1226	18.0	2016	0.1807	0.5075	0.5571	0.9421	[0.07318580904038545, 0.9418258754855011]	[0.1465005669434618, 0.9677246855442564]
0.1226	19.0	2128	0.1801	0.5073	0.5519	0.9449	[0.06989468035151271, 0.9446760689806634]	[0.1326444008392121, 0.9710749826472099]
0.1226	20.0	2240	0.1801	0.5081	0.5539	0.9447	[0.07179866962779094, 0.9444471256159475]	[0.13708319470943298, 0.9707045746832801]
0.1226	21.0	2352	0.1827	0.5072	0.5541	0.9434	[0.07118876107010588, 0.943175745287181]	[0.13893000454912718, 0.9693417271325447]
0.1226	22.0	2464	0.1806	0.5076	0.5528	0.9447	[0.07068067721404604, 0.9444608495376636]	[0.13476280036121427, 0.9707892877994956]
0.1102	23.0	2576	0.1928	0.5083	0.5595	0.9417	[0.07515746582146982, 0.9414232576754142]	[0.15178638113538068, 0.9671506707813141]
0.1102	24.0	2688	0.1886	0.5073	0.5610	0.9396	[0.07519075914020391, 0.9393404945090548]	[0.15723175426565544, 0.9648461896379398]
0.1102	25.0	2800	0.1983	0.5061	0.5622	0.9377	[0.07493822713853636, 0.9373609789235116]	[0.16175032760505428, 0.962676461960256]
0.1102	26.0	2912	0.1882	0.5072	0.5567	0.9419	[0.07270923578891461, 0.9416422211689716]	[0.14593362348164393, 0.9675531811333773]
0.1015	27.0	3024	0.1885	0.5076	0.5550	0.9434	[0.0720122822696466, 0.9431955900992436]	[0.14060537340186446, 0.9693112107227008]
0.1015	28.0	3136	0.1916	0.5081	0.5566	0.9432	[0.0733902558056171, 0.9429066479653452]	[0.14422260848310373, 0.9689043799470217]
0.1015	29.0	3248	0.1938	0.5074	0.5606	0.9400	[0.07503118436885044, 0.9397141501322746]	[0.15591114943543294, 0.9652699739746432]
0.1015	30.0	3360	0.1880	0.5071	0.5450	0.9491	[0.06535303105902995, 0.9489053008277266]	[0.11405408708523163, 0.9759907496995939]

Framework versions

Transformers 4.37.1
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1

Library: transformers
Type: segformer
License: openrail
Created: Jan 24, 2024
Updated: Jan 24, 2024