serdarcaglar's picture
Add SetFit model
a252466 verified
|
raw
history blame
7.66 kB
metadata
library_name: setfit
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
metrics:
  - accuracy
widget:
  - text: >-
      Sarah has 10 stickers. She gives 3 to her friend. What fraction of her
      stickers did Sarah give away?
  - text: >-
      If you have 8 apples and you eat 3 of them, how many apples do you have
      left?
  - text: >-
      What simple strategy could you use to solve this word problem: 'Mike had 9
      candies...'
  - text: >-
      If you have 20 marbles and you give 5 of them to your friend, how many
      marbles do you have left?
  - text: What is the name of the holiday that celebrates workers in September?
pipeline_tag: text-classification
inference: true
base_model: sentence-transformers/all-MiniLM-L6-v2
model-index:
  - name: SetFit with sentence-transformers/all-MiniLM-L6-v2
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: Unknown
          type: unknown
          split: test
        metrics:
          - type: accuracy
            value: 1
            name: Accuracy

SetFit with sentence-transformers/all-MiniLM-L6-v2

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/all-MiniLM-L6-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:

  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 Sources

Model Labels

Label Examples
math
  • 'There are 10 frogs on a log. Some frogs jumped off and now there are 6 frogs left. How can you show this using an equation?'
  • 'Sarah has 9 stickers. She gives 3 stickers to her brother. How many stickers does Sarah have left?'
  • 'Which 3D shape has one curved surface?'
non_math
  • 'What is the currency used in Japan?'
  • 'What do you call a baby kangaroo?'
  • 'What is the capital city of Canada, our neighbor to the north?'

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("serdarcaglar/primary-school-math-question")
# Run inference
preds = model("What is the name of the holiday that celebrates workers in September?")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 3 13.765 33
Label Training Sample Count
math 141
non_math 59

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
  • seed: 42
  • eval_max_steps: -1
  • load_best_model_at_end: False

Training Results

Epoch Step Training Loss Validation Loss
0.002 1 0.3356 -
0.1 50 0.0577 -
0.2 100 0.0053 -
0.3 150 0.0025 -
0.4 200 0.0016 -
0.5 250 0.0008 -
0.6 300 0.0003 -
0.7 350 0.0005 -
0.8 400 0.0006 -
0.9 450 0.0005 -
1.0 500 0.0009 -

Framework Versions

  • Python: 3.10.12
  • SetFit: 1.0.3
  • Sentence Transformers: 2.6.1
  • Transformers: 4.38.2
  • PyTorch: 2.2.1+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}
}