metadata

library_name: setfit
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
metrics:
  - accuracy
widget:
  - text: i honestly thought impossible at this point i feel pretty
  - text: >-
      i feel convinced that im going to shy away from whatever is really good
      for me
  - text: i feel guilt that i should be more caring and im not
  - text: >-
      i found myself feeling nostalgic as i thought about the temporarily
      abandoned little bishop chronicles
  - text: i am feeling very indecisive and spontaneous
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: 0.442
            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: 6 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
3	'im fine mary anne answered feeling a little impatient' 'i feel outraged that my life is so easy so blessed' 'i suggested greys and blues with warm tones as the room is north facing and could feel quite cold and flat'
4	'i still feel its a little shaky at times and can move into the slightly odd jades hair in particular seems prone to this but generally it works well with spencers writing' 'i also feel strange that by the ripe old age of twenty three i want a goddamn life partner' 'i feel acclimated like i am finally a part of this organization rather than a timid observer'
5	'i always feel so flattered when another amazing blogger asks me to share a little of world on their blog so here it goes' 'i was like oh thats awesome blah but then he was like reminding me hes interested in this other girl and i was like i know this but what concerns me more is if it makes you feel too weird to be with me like this' 'i gotta say im feeling pretty impressed with how everything ended up considering my total dollars dropped totaled and i have three small canvases to play with display with'
1	'i still am not able to remember a single dull moment a detail that pissed me off a thing i didnt feel comfortable about' 'i can t find anything to feel other than complacent' 'i am feeling a little bouncy right now'
0	'i hide this secret inside of me away from everyone because i feel ashamed and like i have no assistance in making it better' 'i could feel how exhausted my arms and legs were' 'i am also feeling awful'
2	'i can feel passionate about taking a stand and maybe understand that this one as yet to be chosen issue is worthy of my time and efforts' 'i do awaken from a mild night sweat i usually feel hot as if i had a fever and i want to remove some of my blankets' 'i am feeling quite fond of my friends'

Evaluation

Metrics

Label	Accuracy
all	0.442

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-emotion")
# Run inference
preds = model("i am feeling very indecisive and spontaneous")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	4	18.7292	46

Label	Training Sample Count
0	8
1	8
2	8
3	8
4	8
5	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
0.0042	1	0.2264	-
0.2083	50	0.0997	-
0.4167	100	0.0257	-
0.625	150	0.0089	-
0.8333	200	0.003	-

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}
}