yaniseuranova's picture
Add SetFit model
7f18162 verified
|
raw
history blame
9.34 kB
metadata
library_name: setfit
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
base_model: BAAI/bge-m3
metrics:
  - accuracy
widget:
  - text: >-
      What is the primary difference between a Bayesian neural network and a
      traditional feedforward neural network in the context of machine learning?
  - text: >-
      What is the difference betweensupervised and unsupervised machine learning
      algorithms in terms of data labeling and model training?
  - text: >-
      What is the primary application of Natural Language Processing (NLP) in
      Google's BERT language model, and how does it utilize masked language
      modeling to improve contextual understanding?
  - text: >-
      What is the main advantage of using GraphQL over traditional RESTful APIs,
      as demonstrated by social media giant Facebook in their Facebook ADS API?
  - text: Qui est Robin Mancini ?
pipeline_tag: text-classification
inference: true
model-index:
  - name: SetFit with BAAI/bge-m3
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: Unknown
          type: unknown
          split: test
        metrics:
          - type: accuracy
            value: 1
            name: Accuracy

SetFit with BAAI/bge-m3

This is a SetFit model that can be used for Text Classification. This SetFit model uses BAAI/bge-m3 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:

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

Model Details

Model Description

  • Model Type: SetFit
  • Sentence Transformer body: BAAI/bge-m3
  • Classification head: a LogisticRegression instance
  • Maximum Sequence Length: 8192 tokens
  • Number of Classes: 2 classes

Model Sources

Model Labels

Label Examples
lexical
  • 'What is the definition of semantics in the context ofontology-based data integration, and how does it differ from outright data normalization, as implementented in graph databases like neo4j orAmazon Neptune?'
  • 'What is the primary application of graph convolutional neural networks (GCNNs) in natural language processing (NLP) for modeling syntactic dependencies in parsing?'
  • "What is the distinguising feature of Apache Hive's Metadata Tables, used for maintaining and managingtables in Hadoop Distributed File System (HDFS)?"
semantic
  • 'What is a key challenge faced by managers in sustaining a work culture that encourages creativity, innovation, and critical thinking within the technological industry globally?'
  • 'How might shifting societal values influence the dynamics between multinational corporations and governments, leading to Changes in the global economic landscape?'
  • 'How does the allocation of limited resources affect the allocation of decision-making power within an organization?'

Evaluation

Metrics

Label Accuracy
all 1.0

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("yaniseuranova/setfit-paraphrase-mpnet-base-v2-sst2")
# Run inference
preds = model("Qui est Robin Mancini ?")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 4 19.1392 56
Label Training Sample Count
lexical 36
semantic 43

Training Hyperparameters

  • batch_size: (16, 16)
  • num_epochs: (4, 4)
  • max_steps: -1
  • sampling_strategy: oversampling
  • body_learning_rate: (2e-05, 1e-05)
  • head_learning_rate: 0.01
  • loss: CosineSimilarityLoss
  • distance_metric: cosine_distance
  • margin: 0.25
  • end_to_end: False
  • use_amp: False
  • warmup_proportion: 0.1
  • seed: 42
  • eval_max_steps: -1
  • load_best_model_at_end: True

Training Results

Epoch Step Training Loss Validation Loss
0.0050 1 0.1549 -
0.2475 50 0.0045 -
0.4950 100 0.0009 -
0.7426 150 0.0005 -
0.9901 200 0.0005 -
1.0 202 - 0.0001
1.2376 250 0.0006 -
1.4851 300 0.0006 -
1.7327 350 0.0005 -
1.9802 400 0.0004 -
2.0 404 - 0.0
2.2277 450 0.0003 -
2.4752 500 0.0003 -
2.7228 550 0.0003 -
2.9703 600 0.0003 -
3.0 606 - 0.0
3.2178 650 0.0003 -
3.4653 700 0.0004 -
3.7129 750 0.0003 -
3.9604 800 0.0002 -
4.0 808 - 0.0
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.12
  • SetFit: 1.0.3
  • Sentence Transformers: 2.6.1
  • Transformers: 4.39.0
  • PyTorch: 2.3.0+cu121
  • Datasets: 2.18.0
  • Tokenizers: 0.15.2

Citation

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}
}