You can deploy this model and start earning money today!

usman4d/
setfit-paraphrase-mpnet-base-v2-sst2

Name: setfit-paraphrase-mpnet-base-v2-sst2
Availability: InStock

setfit

safetensors

mpnet

sentence-transformers

text-classification

generated_from_setfit_trainer

2209.11055

sentence-transformers/all-mpnet-base-v2

finetune

model-index

Views

No views yet

SetFit with sentence-transformers/all-mpnet-base-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-mpnet-base-v2 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.

The model has been trained using an efficient few-shot learning technique that involves:

Fine-tuning a Sentence Transformer with contrastive learning.
Training a classification head with features from the fine-tuned Sentence Transformer.

Model Details

Model Description

Model Type: SetFit
Sentence Transformer body: sentence-transformers/all-mpnet-base-v2
Classification head: a LogisticRegression instance
Maximum Sequence Length: 384 tokens
Number of Classes: 10 classes

Model Sources

Repository: SetFit on GitHub
Paper: Efficient Few-Shot Learning Without Prompts
Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

Model Labels

Label	Examples
assumptions_dependencies	"It is assumed that all users will have the necessary hardware, such as smartphones or computers, to access the system's features." 'The system assumes that the external payment gateway (PayPal) will remain operational and accessible for all financial transactions.' 'The software relies on the cloud-based storage solution (AWS S3) for file management, and its availability is critical for proper system operation.'
product_perspective	'This product serves as a front-end interface to an existing backend API and database, providing an improved user experience and additional functionality.' 'This system is a component of a larger enterprise resource planning (ERP) suite and will interact with existing HR and finance modules.' 'The system is part of a multi-tier architecture, serving as the client-side component that communicates with a centralized server via RESTful APIs.'
operating_environment	'The platform will support integration with existing enterprise systems, including SAP for resource management and Salesforce for CRM functionalities.' 'The system will operate on Windows and macOS platforms with support for the latest versions of popular web browsers, including Chrome, Firefox, and Safari.' 'The software will run on desktop devices with a minimum of 4GB RAM and will require a stable internet connection for real-time updates and synchronization.'
user_classes	'Administrators will have full access to system configurations, user management, and reporting tools, and are expected to have technical proficiency.' 'Regular users will be able to browse content, perform basic interactions, and manage their own accounts with minimal training required.' 'Students will interact with the system to view grades, submit assignments, and receive notifications, with a user-friendly interface designed for ease of use.'
design_implementation_constraints	'The software must be able to operate offline for certain use cases, requiring local caching of data and synchronization when the device is online.' 'The mobile app must be optimized for devices with limited processing power, ensuring that it performs efficiently on smartphones with less than 2GB of RAM.' 'The application must be compatible with the existing SQL-based database, and no NoSQL solutions are permitted due to company policy.'
purpose	'The SRS provides a comprehensive description of the Travel Booking System to guide its development and implementation.' 'The purpose of this SRS is to serve as a reference for developers, testers, and stakeholders during the development lifecycle of the CRM solution.' 'The purpose of this document is to define the functional and non-functional requirements of the Student Management System.'
product_functions	'The application will allow users to upload, share, and comment on documents in a collaborative workspace with version control.' 'The platform will provide real-time inventory tracking, including updating stock levels when items are purchased or returned.' 'The system will provide reporting tools for generating sales summaries, customer activity logs, and performance metrics.'
overview	'This document outlines the overall structure and key components of the software system, including its purpose, functionality, and constraints.' 'This specification serves as a foundational reference for the software development team and stakeholders throughout the project lifecycle.' "The SRS provides a comprehensive overview of the software's objectives, user interactions, and integration with external systems."
functional_requirements	'The application shall support multiple payment methods, including credit/debit cards, PayPal, and bank transfers, and process payments securely.' 'The application shall enable users to search for products by name, category, price range, and other filters such as ratings and availability.' 'The software shall allow administrators to add, edit, and delete user accounts, with role-based permissions for different types of users.'
scope	'This software will streamline appointment booking, notification management, and patient records for small to mid-sized clinics.' 'The CRM solution will manage customer interactions, sales pipelines, support tickets, and marketing campaigns in a unified platform.' 'This project includes the development of a cross-platform mobile app for customers to manage their bank accounts, transfer funds, and pay bills.'

Evaluation

Metrics

Label	Accuracy
all	0.875

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

python

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("usman4d/setfit-paraphrase-mpnet-base-v2-sst2")
# Run inference
preds = model("This document aims to capture all the software requirements for the Online Appointment Scheduling System.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	14	20.1094	26

Label	Training Sample Count
assumptions_dependencies	7
design_implementation_constraints	6
functional_requirements	4
operating_environment	8
overview	7
product_functions	8
product_perspective	6
purpose	6
scope	5
user_classes	7

Training Hyperparameters

batch_size: (16, 16)
num_epochs: (1, 1)
max_steps: -1
sampling_strategy: oversampling
num_iterations: 20
body_learning_rate: (2e-05, 2e-05)
head_learning_rate: 2e-05
loss: CosineSimilarityLoss
distance_metric: cosine_distance
margin: 0.25
end_to_end: False
use_amp: False
warmup_proportion: 0.1
l2_weight: 0.01
seed: 42
eval_max_steps: -1
load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0063	1	0.1736	-
0.3125	50	0.1257	-
0.625	100	0.034	-
0.9375	150	0.0177	-

Framework Versions

Python: 3.11.12
SetFit: 1.1.2
Sentence Transformers: 3.4.1
Transformers: 4.50.3
PyTorch: 2.6.0+cu124
Datasets: 3.5.0
Tokenizers: 0.21.1

Citation

BibTeX

bibtex

@article{https://doi.org/10.48550/arxiv.2209.11055,
    doi = {10.48550/ARXIV.2209.11055},
    url = {https://arxiv.org/abs/2209.11055},
    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
    title = {Efficient Few-Shot Learning Without Prompts},
    publisher = {arXiv},
    year = {2022},
    copyright = {Creative Commons Attribution 4.0 International}
}