vit-base-riego Open-source Image Classification Model - Free Deployment for Accurate Image Category Recognition

Vit Base Riego

Developed by MaxP

This model is a fine-tuned version of Google's ViT-base-patch16-224-in21k on an image folder dataset, primarily used for image classification tasks.

Downloads 40

Release Time : 12/30/2022

Model Overview

This is an image classification model based on the Vision Transformer (ViT) architecture, fine-tuned for specific domain image recognition tasks.

Based on ViT Architecture

Utilizes the Vision Transformer architecture, suitable for handling image classification tasks.

Fine-tuned Version

Fine-tuned on a specific dataset based on Google's pre-trained model.

Linear Learning Rate Scheduling

Uses a linear learning rate scheduling strategy during training to optimize training effectiveness.

Image Classification

Visual Feature Extraction

Agricultural Image Analysis

Irrigation System Identification

Identify irrigation systems in agricultural scenes.

Achieved an F1 score of 0.3729 on the test set.

Training Loss	Epoch	Step	Validation Loss	F1
0.1696	0.79	100	1.1385	0.352
0.08	1.59	200	0.9071	0.3774
0.0928	2.38	300	1.1181	0.3454
0.0189	3.17	400	0.8262	0.3425
0.0728	3.97	500	0.9647	0.3747
0.0756	4.76	600	0.6097	0.4776
0.0018	5.56	700	1.3900	0.3652
0.002	6.35	800	0.7498	0.4606
0.0304	7.14	900	1.4367	0.3666
0.0024	7.94	1000	1.5714	0.3041
0.0463	8.73	1100	0.8038	0.4016
0.0014	9.52	1200	0.7175	0.4795
0.0015	10.32	1300	1.0347	0.3959
0.0009	11.11	1400	1.3881	0.3670
0.0131	11.9	1500	1.0780	0.4044
0.0007	12.7	1600	0.9834	0.4255
0.0011	13.49	1700	1.0753	0.4033
0.0007	14.29	1800	1.1514	0.3989
0.0007	15.08	1900	1.2373	0.3769
0.0007	15.87	2000	1.2998	0.3729

Property	Details
Model Type	Fine - tuned version of google/vit - base - patch16 - 224 - in21k
Training Data	imagefolder
Metrics	F1

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base