mit-b4-finetuned-stroke-binary

BTX24/mit-b4-finetuned-stroke-binary

mit-b4-finetuned-stroke-binary is a image classification model from BTX24. Use it when you need a label for an image. 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

8% of all-time downloads

All-time downloads

143

Public

Parameters

60.8M

4.9 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors243 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: other
Model type: segformer
Access: Public
Created: Mar 16, 2025
Updated: Mar 16, 2025
SHA: 388ac6f1

Base models

nvidia/mit-b4

Model card

From the Hugging Face model README

mit-b4-finetuned-stroke-binary

This model is a fine-tuned version of nvidia/mit-b4 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1162
Accuracy: 0.9701
F1: 0.9701
Precision: 0.9701
Recall: 0.9701

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: 2e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 12
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.5714	0.6202	100	0.4776	0.7879	0.7800	0.7900	0.7879
0.3897	1.2357	200	0.3239	0.8716	0.8704	0.8711	0.8716
0.2951	1.8558	300	0.3120	0.8765	0.8724	0.8858	0.8765
0.23	2.4713	400	0.1994	0.9281	0.9271	0.9304	0.9281
0.2135	3.0868	500	0.2157	0.9281	0.9267	0.9333	0.9281
0.2106	3.7070	600	0.1809	0.9380	0.9382	0.9387	0.9380
0.1576	4.3225	700	0.1629	0.9403	0.9404	0.9404	0.9403
0.1434	4.9426	800	0.1526	0.9543	0.9542	0.9543	0.9543
0.1391	5.5581	900	0.1268	0.9575	0.9575	0.9575	0.9575
0.1048	6.1736	1000	0.1489	0.9557	0.9555	0.9558	0.9557
0.1271	6.7938	1100	0.1448	0.9570	0.9566	0.9586	0.9570
0.091	7.4093	1200	0.1451	0.9570	0.9567	0.9580	0.9570
0.1159	8.0248	1300	0.1205	0.9629	0.9627	0.9636	0.9629
0.1151	8.6450	1400	0.1124	0.9665	0.9664	0.9666	0.9665
0.0735	9.2605	1500	0.1175	0.9643	0.9641	0.9645	0.9643
0.0537	9.8806	1600	0.1154	0.9679	0.9678	0.9679	0.9679
0.0666	10.4961	1700	0.1162	0.9701	0.9701	0.9701	0.9701
0.0732	11.1116	1800	0.1133	0.9679	0.9678	0.9679	0.9679
0.0775	11.7318	1900	0.1130	0.9683	0.9683	0.9684	0.9683

Framework versions

Transformers 4.48.3
Pytorch 2.6.0+cu124
Datasets 3.4.0
Tokenizers 0.21.0

image/png

Task: Image Classification
Library: transformers
Type: segformer
License: other
Created: Mar 16, 2025
Updated: Mar 16, 2025