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: 2 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
0	'Which price points to play in?' 'What are some whitespaces in terms of price bracket for Jumex in TT HM CSD?' 'What are the key drivers of growth for kof in ncb?'
1	'what is ROI trend for Fizzy drinks?' 'Give the price vs volume comparison' 'What is the change in industry mix for coca-cola in TT HM Orizaba in 2021 to 2022'

Evaluation

Metrics

Label	Accuracy
all	0.8333

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("vgarg/query_type_classifier_int-e5-large_v2")
# Run inference
preds = model("How has the csd industry evolved in the last two years?")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	5	13.3684	32

Label	Training Sample Count
0	40
1	36

Training Hyperparameters

batch_size: (16, 16)
num_epochs: (3, 3)
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
0.0053	1	0.3156	-
0.2632	50	0.1347	-
0.5263	100	0.0013	-
0.7895	150	0.0003	-
1.0526	200	0.0002	-
1.3158	250	0.0002	-
1.5789	300	0.0001	-
1.8421	350	0.0002	-
2.1053	400	0.0001	-
2.3684	450	0.0001	-
2.6316	500	0.0001	-
2.8947	550	0.0001	-

Framework Versions

Python: 3.12.2
SetFit: 1.0.3
Sentence Transformers: 2.6.1
Transformers: 4.39.3
PyTorch: 2.2.2+cpu
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}
}

vgarg
/

query_type_classifier_int-e5-large_v2