metadata

library_name: setfit
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
metrics:
  - accuracy
widget:
  - text: >-
      we get some truly unique character studies and a cross-section of
      americana that hollywood could n't possibly fictionalize and be believed .
  - text: >-
      the movie is one of the best examples of artful large format filmmaking
      you are likely to see anytime soon .
  - text: my response to the film is best described as lukewarm .
  - text: >-
      the movie 's ripe , enrapturing beauty will tempt those willing to probe
      its inscrutable mysteries .
  - text: >-
      fear dot com is so rambling and disconnected it never builds any suspense
      .
pipeline_tag: text-classification
inference: true
base_model: sentence-transformers/paraphrase-mpnet-base-v2
model-index:
  - name: SetFit with sentence-transformers/paraphrase-mpnet-base-v2
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: Unknown
          type: unknown
          split: test
        metrics:
          - type: accuracy
            value: 0.4447963800904977
            name: Accuracy

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

This is a SetFit model that can be used for Text Classification. This SetFit model uses sentence-transformers/paraphrase-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/paraphrase-mpnet-base-v2
Classification head: a LogisticRegression instance
Maximum Sequence Length: 512 tokens
Number of Classes: 5 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
2	"a farce of a parody of a comedy of a premise , it is n't a comparison to reality so much as it is a commentary about our knowledge of films ." 'mild , meandering teen flick .' 'wanker goths are on the loose !'
4	"the first tunisian film i have ever seen , and it 's also probably the most good-hearted yet sensual entertainment i 'm likely to see all year ." "the film 's intimate camera work and searing performances pull us deep into the girls ' confusion and pain as they struggle tragically to comprehend the chasm of knowledge that 's opened between them ." 'the film is small in scope , yet perfectly formed .'
0	'priggish , lethargically paced parable of renewal .' "not at all clear what it 's trying to say and even if it were -- i doubt it would be all that interesting ." 'an unfortunate title for a film that has nothing endearing about it .'
3	"you may leave the theater with more questions than answers , but darned if your toes wo n't still be tapping ." 'an alternately fascinating and frustrating documentary .' 'instead of hitting the audience over the head with a moral , schrader relies on subtle ironies and visual devices to convey point of view .'
1	"it 's not so much a movie as a joint promotion for the national basketball association and teenaged rap and adolescent poster-boy lil ' bow wow ." 'just the sort of lazy tearjerker that gives movies about ordinary folk a bad name .' 'well , this movie proves you wrong on both counts .'

Evaluation

Metrics

Label	Accuracy
all	0.4448

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("vidhi0206/setfit-paraphrase-mpnet-sst5")
# Run inference
preds = model("my response to the film is best described as lukewarm .")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	2	20.45	43

Label	Training Sample Count
0	8
1	8
2	8
3	8
4	8

Training Hyperparameters

batch_size: (8, 8)
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
0.005	1	0.1967	-
0.25	50	0.0658	-
0.5	100	0.0883	-
0.75	150	0.0385	-
1.0	200	0.0112	-

Framework Versions

Python: 3.8.10
SetFit: 1.0.3
Sentence Transformers: 2.3.1
Transformers: 4.37.2
PyTorch: 2.2.0+cu121
Datasets: 2.17.0
Tokenizers: 0.15.1

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}
}