Levit-192 Fine-tuned Open-source Model - Optimized Based on Unlabeled Datasets with Excellent Performance in Precision and Other Metrics

Levit 192 Finetuned On Unlabelled IA With Snorkel Labels

Developed by ImageIN

This model is a fine-tuned version of facebook/levit-192 on an unlabeled dataset, demonstrating excellent performance in precision, recall, F1 score, and accuracy.

Image Classification

Transformers

Open Source License:Apache-2.0 #High-precision Image Classification #Weakly Supervised Learning #Lightweight Vision Model

Downloads 19

Release Time : 10/9/2022

Model Overview

This vision model is based on the LeViT-192 architecture and fine-tuned using unlabeled data and snorkel labeling techniques, suitable for tasks such as image classification.

Model Features

High-precision Classification

Achieves 98.36% precision and 98.22% recall on the evaluation set.

Unsupervised Fine-tuning

Uses unlabeled data and snorkel labeling techniques for model fine-tuning.

Efficient Training

Optimizes the training process with mixed-precision training and linear learning rate scheduling.

Model Capabilities

Image Classification

Visual Feature Extraction

Use Cases

Computer Vision

General Image Classification

Classifies and recognizes various types of images.

Achieves 98.73% accuracy on the test set.

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	253	nan	0.9743	0.9791	0.9766	0.9826
0.0557	2.0	506	nan	0.9829	0.9801	0.9815	0.9863
0.0557	3.0	759	nan	0.9836	0.9822	0.9829	0.9873
0.0543	4.0	1012	nan	0.9839	0.9775	0.9807	0.9858
0.0543	5.0	1265	nan	0.9616	0.9727	0.9670	0.9752
0.0457	6.0	1518	nan	0.9563	0.9699	0.9629	0.9720
0.0457	7.0	1771	nan	0.9822	0.9808	0.9815	0.9863
0.0418	8.0	2024	nan	0.9735	0.9769	0.9752	0.9815
0.0418	9.0	2277	nan	0.9832	0.9811	0.9822	0.9868
0.0396	10.0	2530	nan	0.9843	0.9815	0.9829	0.9873

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

Levit 192 Finetuned On Unlabelled IA With Snorkel Labels

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 levit-192_finetuned_on_unlabelled_IA_with_snorkel_labels

🚀 Quick Start

📄 License

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions