segformer-b0-doclaynet-full-v1.2

Darayut/segformer-b0-doclaynet-full-v1.2

segformer-b0-doclaynet-full-v1.2 is a machine learning model from Darayut. 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.

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

Downloads · 30 days

7% of all-time downloads

All-time downloads

178

Public

Parameters

3.7M

8.1 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt29.9 MB · 50%

At a glance

Library: transformers
Model type: segformer
Access: Public
Created: Jan 28, 2026
Updated: Jan 29, 2026
SHA: 9f35ebdf

Model card

From the Hugging Face model README

segformer-b0-doclaynet-full-v1.2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2087
Mean Iou: 0.3709

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: 4e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 1000
training_steps: 50000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou
0.2185	0.02	1000	0.2426	0.3607
0.2133	0.04	2000	0.2391	0.3635
0.2165	0.06	3000	0.2462	0.3961
0.2147	0.08	4000	0.2422	0.3538
0.1741	0.1	5000	0.2457	0.3573
0.1638	0.12	6000	0.2435	0.3530
0.2154	0.14	7000	0.2420	0.3586
0.2391	0.16	8000	0.2448	0.3301
0.212	0.18	9000	0.2432	0.3293
0.1971	0.2	10000	0.2399	0.3619
0.1856	0.22	11000	0.2389	0.4055
0.2026	0.24	12000	0.2356	0.3626
0.2022	0.26	13000	0.2417	0.3581
0.2144	0.28	14000	0.2325	0.3558
0.2192	0.3	15000	0.2319	0.5201
0.1874	0.32	16000	0.2362	0.3623
0.2089	0.34	17000	0.2290	0.3595
0.2212	0.36	18000	0.2285	0.3640
0.2061	0.38	19000	0.2357	0.4068
0.2072	0.4	20000	0.2289	0.4024
0.2026	0.42	21000	0.2324	0.4085
0.2036	0.44	22000	0.2327	0.4068
0.1547	0.46	23000	0.2320	0.3682
0.1671	0.48	24000	0.2154	0.4041
0.2551	0.5	25000	0.2200	0.3684
0.2078	0.52	26000	0.2226	0.4078
0.1977	0.54	27000	0.2326	0.3679
0.2152	0.56	28000	0.2183	0.3615
0.1882	0.58	29000	0.2277	0.4073
0.1804	0.6	30000	0.2323	0.3576
0.2055	0.62	31000	0.2199	0.3635
0.2157	0.64	32000	0.2254	0.3650
0.1841	0.66	33000	0.2179	0.3677
0.1843	0.68	34000	0.2161	0.3626
0.2018	0.7	35000	0.2129	0.3684
0.1965	0.72	36000	0.2144	0.3708
0.1924	0.74	37000	0.2111	0.3717
0.1923	0.76	38000	0.2109	0.3728
0.189	0.78	39000	0.2120	0.4093
0.1462	0.8	40000	0.2104	0.3687
0.1471	0.82	41000	0.2055	0.4133
0.1912	0.84	42000	0.2071	0.3709
0.2341	0.86	43000	0.2097	0.3737
0.1892	0.88	44000	0.2072	0.3385
0.1769	0.9	45000	0.2074	0.3720
0.209	0.92	46000	0.2093	0.3711
0.182	0.94	47000	0.2092	0.3712
0.1799	0.96	48000	0.2088	0.3707
0.1844	0.98	49000	0.2080	0.3713
0.1937	1.0	50000	0.2087	0.3709

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu126
Datasets 4.4.2
Tokenizers 0.22.1

Library: transformers
Type: segformer
Created: Jan 28, 2026
Updated: Jan 29, 2026