FewShotIssueClassifier-NLBSE23 Open-source Model - Free Deployment for Precise Classification of 4 Major Categories of Problem Reports

Fewshotissueclassifier NLBSE23

Developed by PeppoCola

A sentence similarity model based on Sentence Transformers, fine-tuned for issue report classification tasks, supporting 4 categories: defect/documentation/feature/question

Text Classification

PyTorch

EnglishOpen Source License:Gpl-3.0 #Issue Report Classification #Few-shot Learning #Sentence Similarity

Downloads 27

Release Time : 3/21/2023

Model Overview

This model maps sentences and paragraphs to a 768-dimensional dense vector space, specifically designed for issue report classification tasks, suitable for scenarios like clustering or semantic search

Model Features

Few-shot Learning Capability

Optimized for small-scale labeled data, suitable for scenarios with limited annotation resources

Multi-category Classification

Supports identification of 4 types of issue reports: defect/documentation/feature/question

High-dimensional Semantic Encoding

Maps text to a 768-dimensional dense vector space, preserving rich semantic information

Model Capabilities

Text Classification

Semantic Similarity Calculation

Automatic Issue Report Categorization

Short Text Vectorization

Use Cases

Software Development Support

Automatic Issue Report Classification

Automatically categorizes user-submitted issue reports into predefined categories

Reduces manual classification workload and improves issue tracking efficiency

Defect Report Analysis

Identifies genuine defect reports from a large number of issue reports

Helps development teams prioritize critical issues

Text Analysis

Semantic Clustering

Automatically groups similar issue reports

Identifies duplicate issues or related requirements

🚀 FewShotIssueClassifier-NLBSE23

This is a SetFit model that utilizes Sentence Transformers to map sentences and paragraphs into a 768-dimensional dense vector space. It can be applied to tasks such as clustering or semantic search.

🚀 Quick Start

This specific model is fine-tuned for Issue Report Classification in 4 classes: bug, documentation, feature, question.

💻 Usage Examples

Basic Usage

from sentence_transformers.losses import CosineSimilarityLoss
from setfit import SetFitModel
from setfit import SetFitTrainer
sentences = ["error in line 20", "add method list_features"]

label_mapping = {
  0 : "bug",
  1 : "documentation",
  2 : "feature",
  3 : "question"
}

model = SetFitModel.from_pretrained('PeppoCola/FewShotIssueClassifier-NLBSE23')
predictions = model.predict(sentences)
print([label_mapping[i] for i in predictions])

📚 Documentation

Dataset

This model is trained on a subset of the NLBSE23 dataset. The sample was hand-labeled, and made available on Zenodo.

Citing & Authors

@software{Colavito_Few-Shot_Learning_for_2023,
	title        = {{Few-Shot Learning for Issue Report Classification}},
	author       = {Colavito, Giuseppe and Lanubile, Filippo and Novielli, Nicole},
	year         = 2023,
	month        = 2,
	url          = {https://github.com/collab-uniba/Issue-Report-Classification-NLBSE2023},
	version      = {1.0.0}
}

@dataset{colavito_giuseppe_2023_7628150,
  author       = {Colavito Giuseppe and
                  Lanubile Filippo and
                  Novielli Nicole},
  title        = {Few-Shot Learning for Issue Report Classification},
  month        = feb,
  year         = 2023,
  note         = {{To use this, merge the CSV with the original 
                   dataset (after removing duplicates on the 'id'
                   column)}},
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.7628150},
  url          = {https://doi.org/10.5281/zenodo.7628150}
}

@inproceedings{Colavito-2023,
	title        = {Few-Shot Learning for Issue Report Classification},
	author       = {Colavito, Giuseppe and Lanubile, Filippo and Novielli, Nicole},
	year         = 2023,
	booktitle    = {2nd International Workshop on Natural Language-Based Software Engineering (NLBSE)}
}

📄 License

This model is released under the GPL-3.0 license.

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