Open-source Image Classification Model exper_batch_8_e4 - Free Deployment for Precise Image Category Recognition

Exper Batch 8 E4

Developed by sudo-s

An image classification model fine-tuned on the sudo-s/herbier_mesuem1 dataset based on google/vit-base-patch16-224-in21k

Image Classification

Transformers

Open Source License:Apache-2.0 #Plant Image Classification #High Accuracy #ViT Fine-tuning

Downloads 31

Release Time : 6/26/2022

Model Overview

This is a Vision Transformer model specifically designed for image classification tasks. It has been fine-tuned on a specific dataset, achieving an accuracy of 91.83%.

Model Features

High Accuracy

Achieved a classification accuracy of 91.83% on the evaluation set

Based on ViT Architecture

Utilizes the Vision Transformer base architecture with powerful image feature extraction capabilities

Efficient Fine-tuning

Requires only 4 training epochs to achieve good performance

Model Capabilities

Image Classification

Feature Extraction

Use Cases

Plant Identification

Plant Specimen Classification

Can be used for automatic classification of plant specimens in museums or research institutions

Accuracy reached 91.83%

Educational Applications

Botany Teaching Aid

Helps students identify different plant species

🚀 exper_batch_8_e4

This model is a fine - tuned version of [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k) on the sudo - s/herbier_mesuem1 dataset. It offers high - performance image classification capabilities, achieving excellent results on the evaluation set.

🚀 Quick Start

This model is a fine - tuned version of [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k) on the sudo - s/herbier_mesuem1 dataset. It achieves the following results on the evaluation set:

Loss: 0.3353
Accuracy: 0.9183

📚 Documentation

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: 0.0002
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
num_epochs: 4
mixed_precision_training: Apex, opt level O1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
4.2251	0.08	100	4.1508	0.1203
3.4942	0.16	200	3.5566	0.2082
3.2871	0.23	300	3.0942	0.3092
2.7273	0.31	400	2.8338	0.3308
2.4984	0.39	500	2.4860	0.4341
2.3423	0.47	600	2.2201	0.4796
1.8785	0.55	700	2.1890	0.4653
1.8012	0.63	800	1.9901	0.4865
1.7236	0.7	900	1.6821	0.5736
1.4949	0.78	1000	1.5422	0.6083
1.5573	0.86	1100	1.5436	0.6110
1.3241	0.94	1200	1.4077	0.6207
1.0773	1.02	1300	1.1417	0.6916
0.7935	1.1	1400	1.1194	0.6931
0.7677	1.17	1500	1.0727	0.7167
0.9468	1.25	1600	1.0707	0.7136
0.7563	1.33	1700	0.9427	0.7390
0.8471	1.41	1800	0.8906	0.7571
0.9998	1.49	1900	0.8098	0.7845
0.6039	1.57	2000	0.7244	0.8034
0.7052	1.64	2100	0.7881	0.7953
0.6753	1.72	2200	0.7458	0.7926
0.3758	1.8	2300	0.6987	0.8022
0.4985	1.88	2400	0.6286	0.8265
0.4122	1.96	2500	0.5949	0.8358
0.1286	2.04	2600	0.5691	0.8385
0.1989	2.11	2700	0.5535	0.8389
0.3304	2.19	2800	0.5261	0.8520
0.3415	2.27	2900	0.5504	0.8477
0.4066	2.35	3000	0.5418	0.8497
0.1208	2.43	3100	0.5156	0.8612
0.1668	2.51	3200	0.5655	0.8539
0.0727	2.58	3300	0.4971	0.8658
0.0929	2.66	3400	0.4962	0.8635
0.0678	2.74	3500	0.4903	0.8670
0.1212	2.82	3600	0.4357	0.8867
0.1579	2.9	3700	0.4642	0.8739
0.2625	2.98	3800	0.3994	0.8951
0.024	3.05	3900	0.3953	0.8971
0.0696	3.13	4000	0.3883	0.9056
0.0169	3.21	4100	0.3755	0.9086
0.023	3.29	4200	0.3685	0.9109
0.0337	3.37	4300	0.3623	0.9109
0.0123	3.45	4400	0.3647	0.9067
0.0159	3.52	4500	0.3630	0.9082
0.0154	3.6	4600	0.3522	0.9094
0.0112	3.68	4700	0.3439	0.9163
0.0219	3.76	4800	0.3404	0.9194
0.0183	3.84	4900	0.3371	0.9183
0.0103	3.92	5000	0.3362	0.9183
0.0357	3.99	5100	0.3353	0.9183

Framework versions

Transformers 4.19.4
Pytorch 1.5.1
Datasets 2.3.2
Tokenizers 0.12.1

📄 License

This project is licensed under the Apache - 2.0 license.

Property	Details
Model Type	Fine - tuned version of [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k)
Training Data	sudo - s/herbier_mesuem1 dataset

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご