segformer-b1-finetuned-segments-pv_v1_normalized_p100_4batch_fp

mouadenna/segformer-b1-finetuned-segments-pv_v1_normalized_p100_4batch_fp

segformer-b1-finetuned-segments-pv_v1_normalized_p100_4batch_fp is a image segmentation model from mouadenna. 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

21% of all-time downloads

All-time downloads

Public

Parameters

13.7M

54.8 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors54.7 MB · 100%

At a glance

Task: Image Segmentation
Library: transformers
License: other
Model type: segformer
Access: Public
Created: Jul 28, 2024
Updated: Jul 28, 2024
SHA: ddb6290a

Base models

nvidia/segformer-b1-finetuned-ade-512-512

Model card

From the Hugging Face model README

segformer-b1-finetuned-segments-pv_v1_normalized_p100_4batch_fp

This model is a fine-tuned version of nvidia/segformer-b1-finetuned-ade-512-512 on the mouadenna/satellite_PV_dataset_train_test_v1 dataset. It achieves the following results on the evaluation set:

Loss: 0.0012
Mean Iou: 0.9589
Precision: 0.9794

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: 0.0004
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.001
num_epochs: 40
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Precision
0.0641	0.9989	229	0.0082	0.8288	0.8881
0.0077	1.9978	458	0.0070	0.8228	0.8650
0.0058	2.9967	687	0.0042	0.8827	0.9339
0.005	4.0	917	0.0039	0.8849	0.9172
0.0044	4.9989	1146	0.0071	0.7938	0.8122
0.0049	5.9978	1375	0.0036	0.8914	0.9402
0.0045	6.9967	1604	0.0042	0.8729	0.9280
0.0038	8.0	1834	0.0035	0.8889	0.9433
0.0034	8.9989	2063	0.0030	0.9038	0.9357
0.0032	9.9978	2292	0.0026	0.9115	0.9501
0.003	10.9967	2521	0.0026	0.9136	0.9482
0.0031	12.0	2751	0.0026	0.9132	0.9461
0.0029	12.9989	2980	0.0026	0.9144	0.9493
0.0026	13.9978	3209	0.0023	0.9202	0.9414
0.0025	14.9967	3438	0.0024	0.9175	0.9456
0.003	16.0	3668	0.0032	0.8926	0.9640
0.0035	16.9989	3897	0.0041	0.8741	0.9007
0.0029	17.9978	4126	0.0022	0.9229	0.9598
0.0024	18.9967	4355	0.0022	0.9239	0.9549
0.0022	20.0	4585	0.0020	0.9308	0.9601
0.0021	20.9989	4814	0.0019	0.9325	0.9689
0.0021	21.9978	5043	0.0019	0.9334	0.9630
0.002	22.9967	5272	0.0018	0.9368	0.9631
0.002	24.0	5502	0.0019	0.9333	0.9684
0.002	24.9989	5731	0.0018	0.9381	0.9613
0.0022	25.9978	5960	0.0018	0.9369	0.9610
0.0019	26.9967	6189	0.0017	0.9413	0.9677
0.0018	28.0	6419	0.0016	0.9429	0.9629
0.0017	28.9989	6648	0.0016	0.9444	0.9642
0.0017	29.9978	6877	0.0015	0.9465	0.9741
0.0016	30.9967	7106	0.0014	0.9492	0.9718
0.0016	32.0	7336	0.0014	0.9499	0.9687
0.0015	32.9989	7565	0.0015	0.9469	0.9737
0.0016	33.9978	7794	0.0014	0.9514	0.9721
0.0015	34.9967	8023	0.0013	0.9542	0.9719
0.0014	36.0	8253	0.0013	0.9546	0.9694
0.0014	36.9989	8482	0.0012	0.9569	0.9740
0.0014	37.9978	8711	0.0012	0.9579	0.9781
0.0014	38.9967	8940	0.0012	0.9584	0.9759
0.0013	39.9564	9160	0.0012	0.9589	0.9794

Framework versions

Transformers 4.42.3
Pytorch 2.1.2
Datasets 2.20.0
Tokenizers 0.19.1

Task: Image Segmentation
Library: transformers
Type: segformer
License: other
Created: Jul 28, 2024
Updated: Jul 28, 2024