segformer-b1-finetuned-segments-pv_v1_normalized_t4_4batch_augx3

mouadenna/segformer-b1-finetuned-segments-pv_v1_normalized_t4_4batch_augx3

segformer-b1-finetuned-segments-pv_v1_normalized_t4_4batch_augx3 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

7% of all-time downloads

All-time downloads

Public

Parameters

13.7M

57.7 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors54.7 MB · 95%

At a glance

Task: Image Segmentation
Library: transformers
License: other
Model type: segformer
Access: Public
Created: Jul 26, 2024
Updated: Jul 26, 2024
SHA: 35a7e6ae

Base models

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

Model card

From the Hugging Face model README

segformer-b1-finetuned-segments-pv_v1_normalized_t4_4batch_augx3

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: 0.0111
Mean Iou: 0.8679
Precision: 0.9181

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 40
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Precision
0.0086	0.9978	343	0.0064	0.8154	0.9105
0.0031	1.9985	687	0.0058	0.8290	0.9324
0.0059	2.9993	1031	0.0061	0.8077	0.9352
0.0031	4.0	1375	0.0057	0.8347	0.9182
0.0043	4.9978	1718	0.0059	0.8275	0.8747
0.0034	5.9985	2062	0.0054	0.8499	0.8922
0.0035	6.9993	2406	0.0055	0.8508	0.9072
0.0037	8.0	2750	0.0055	0.8466	0.9201
0.0028	8.9978	3093	0.0054	0.8590	0.9139
0.0031	9.9985	3437	0.0055	0.8562	0.9135
0.002	10.9993	3781	0.0054	0.8619	0.9175
0.0019	12.0	4125	0.0056	0.8649	0.9131
0.0023	12.9978	4468	0.0061	0.8632	0.9195
0.0028	13.9985	4812	0.0061	0.8553	0.9174
0.0041	14.9993	5156	0.0072	0.8573	0.9172
0.002	16.0	5500	0.0063	0.8643	0.9136
0.0019	16.9978	5843	0.0068	0.8637	0.9185
0.0019	17.9985	6187	0.0073	0.8598	0.9123
0.0015	18.9993	6531	0.0070	0.8620	0.9108
0.0019	20.0	6875	0.0073	0.8602	0.9163
0.0017	20.9978	7218	0.0071	0.8669	0.9229
0.0014	21.9985	7562	0.0081	0.8633	0.9198
0.0027	22.9993	7906	0.0089	0.8573	0.9138
0.0017	24.0	8250	0.0086	0.8570	0.9114
0.0011	24.9978	8593	0.0087	0.8635	0.9152
0.0013	25.9985	8937	0.0100	0.8583	0.9203
0.0012	26.9993	9281	0.0085	0.8651	0.9113
0.0015	28.0	9625	0.0091	0.8697	0.9179
0.0011	28.9978	9968	0.0091	0.8684	0.9204
0.0012	29.9985	10312	0.0099	0.8658	0.9152
0.001	30.9993	10656	0.0098	0.8663	0.9170
0.0011	32.0	11000	0.0100	0.8680	0.9174
0.0008	32.9978	11343	0.0102	0.8675	0.9181
0.0009	33.9985	11687	0.0107	0.8669	0.9180
0.0011	34.9993	12031	0.0107	0.8681	0.9214
0.001	36.0	12375	0.0113	0.8677	0.9166
0.001	36.9978	12718	0.0115	0.8669	0.9179
0.001	37.9985	13062	0.0109	0.8694	0.9206
0.0008	38.9993	13406	0.0112	0.8681	0.9175
0.0009	39.9127	13720	0.0111	0.8679	0.9181

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 26, 2024
Updated: Jul 26, 2024