metadata

library_name: setfit
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
base_model: BAAI/bge-small-en-v1.5
metrics:
  - accuracy
widget:
  - text: >-
      Can you tell I about eny ongoing promoistion onr discounts onteh organic
      produce?
  - text: >-
      A bought somenting that didn ' th meet my expectations. It there ein way
      go get and partial refund?
  - text: >-
      I ' d like to palac a ladge ordet for my business. Do you offer ang
      specialy bulk shopping rates?
  - text: >-
      Ken you telle mo more about the origin atch farming practices of your
      cofffee beans?
  - text: >-
      I ' d llike to exchange a product I bought in - store. Du hi needs yo
      bring tie oringal receipt?
pipeline_tag: text-classification
inference: true
model-index:
  - name: SetFit with BAAI/bge-small-en-v1.5
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: Unknown
          type: unknown
          split: test
        metrics:
          - type: accuracy
            value: 0.9056603773584906
            name: Accuracy

SetFit with BAAI/bge-small-en-v1.5

This is a SetFit model that can be used for Text Classification. This SetFit model uses BAAI/bge-small-en-v1.5 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: BAAI/bge-small-en-v1.5
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
Tech Support	"I ' am trying to place an orden online bt Then website keeps crashing. Can you assit my?" "Mi online order won ' t go throw - is there an isuue with years pament prossesing?" "I ' m goning an error when tryied tou redeem my loyality points. Who cen assist we?"
HR	"I ' m considere submitting my ow - weeck notice. Waht It's tehe typical resignation process?" "I ' m looking e swich to a part - time sehdule. Whate re rhe requirements?" "In ' d loke to fill a formal complain about worksplace discrimination. Who did I contact?"
Product	'Whots are your best practices ofr mantain foord quality and freshness?' 'Whots newbrand ow nut butters dou you carry tahat are peanut - free?' 'Do you hafe any seasonal nor limited - tíme produts in stock rignt now?'
Returns	'My grocery delivary contained items tath where spoiled or pas their expiration date. How dos me get replacements?' "I ' d llike to exchange a product I bought in - store. Du hi needs yo bring tie oringal receipt?" 'I eceibed de demaged item in my online oder. Hou do I’m go about getting a refund?'
Logistics	'I have a question about youtr Holiday shiping deathlines and prioritized delivery options' 'I nedd to change the delivery addrss foy mh upcoming older. How can I go that?' 'Can jou explain York polices around iterms that approxmatlly out of stock or on backorder?'

Evaluation

Metrics

Label	Accuracy
all	0.9057

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("setfit_model_id")
# Run inference
preds = model("Can you tell I about eny ongoing promoistion onr discounts onteh organic produce?")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	10	16.125	28

Label	Training Sample Count
Returns	8
Tech Support	8
Logistics	8
HR	8
Product	8

Training Hyperparameters

batch_size: (32, 32)
num_epochs: (10, 10)
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: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.025	1	0.2185	-
1.25	50	0.0888	-
2.5	100	0.0157	-
3.75	150	0.0053	-
5.0	200	0.0033	-
6.25	250	0.004	-
7.5	300	0.0024	-
8.75	350	0.0027	-
10.0	400	0.0025	-

Framework Versions

Python: 3.11.8
SetFit: 1.0.3
Sentence Transformers: 2.6.1
Transformers: 4.39.3
PyTorch: 2.4.0.dev20240413
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}
}