metadata

library_name: setfit
tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
base_model: intfloat/multilingual-e5-large
metrics:
  - accuracy
widget:
  - text: >-
      What promotions in RTEC have shown declining effectiveness and can be
      discontinued?
  - text: What are my priority brands in RTEC to get positive Lift Change in 2022?
  - text: >-
      What would be the expected incremental volume lift if the discount on
      Brand Zucaritas is raised by 5%?
  - text: Which promotion types are better for low discounts for Zucaritas ?
  - text: Which Promotions contributred the most ROI Change between 2022 and 2023?
pipeline_tag: text-classification
inference: true
model-index:
  - name: SetFit with intfloat/multilingual-e5-large
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: Unknown
          type: unknown
          split: test
        metrics:
          - type: accuracy
            value: 0.9714285714285714
            name: Accuracy

SetFit with intfloat/multilingual-e5-large

This is a SetFit model that can be used for Text Classification. This SetFit model uses intfloat/multilingual-e5-large 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: intfloat/multilingual-e5-large
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
2	'Which brand has the highest change in lift for NATURAL JUICES category in 2022?' 'What are the main reasons for Lift decline for ULTRASTORE in 2023 compared to 2022?' 'Why has the overall Lift declined in 2023 in BREEZEFIZZ vs 2022?'
5	'How will the introduction of a 20% discount promotion for Rice Krispies in August affect incremental volume and ROI?' 'If I raise the discount by 20% on Brand BREEZEFIZZ, what will be the incremental roi?' 'How will increasing the discount by 50 percent on Brand BREEZEFIZZ affect the incremental volume lift?'
1	'How do the performance metrics of brands in the FIZZY DRINKS category compare to those in HYDRA and NATURAL JUICES concerning ROI change between 2021 to 2022?' 'Were there any sku-specific promotions that may have influenced their ROI and contributed to the overall decline?' 'Which category has contributed the most to ROI change between 2021 to 2022?'
0	'How is the promotion efficacy in 2022 compared to 2021 for CHEDRAUI channel?' 'Which subcategory have the highest ROI in 2022?' 'Which channel has the max ROI and Vol Lift when we run the Promotion for FIZZY DRINKS category?'
3	'Which promotion types are better for high discounts in Hydra category for 2022?' 'Which promotion types are preferable for high discounts in FIZZY DRINKS for CORN POPS?' 'Which promotion strategies in FIZZY DRINKS allow for offering substantial discounts while maintaining profitability?'
4	'Which promotions have scope for higher investment to drive more ROIs in Hydra ?' 'How can Hydra category investors diversify their investment portfolio to improve ROI?' 'For FIZZY DRINKS what would be a better investment strategy to gain ROI'

Evaluation

Metrics

Label	Accuracy
all	0.9714

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/promo_prescriptive_gpt_30_04_2024_v1")
# Run inference
preds = model("Which promotion types are better for low discounts for Zucaritas ?")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	8	15.1667	27

Label	Training Sample Count
0	10
1	10
2	10
3	10
4	10
5	10

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
0.0067	1	0.3577	-
0.3333	50	0.04	-
0.6667	100	0.002	-
1.0	150	0.0013	-
1.3333	200	0.0009	-
1.6667	250	0.0006	-
2.0	300	0.0006	-
2.3333	350	0.0004	-
2.6667	400	0.0006	-
3.0	450	0.0004	-

Framework Versions

Python: 3.10.12
SetFit: 1.0.3
Sentence Transformers: 2.7.0
Transformers: 4.40.1
PyTorch: 2.2.1+cu121
Datasets: 2.19.0
Tokenizers: 0.19.1

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}
}