Hugging Face's logo

Shankhdhar
/
classifier_woog_arista_vault_combined

Text Classification
setfit
Safetensors
sentence-transformers
mpnet
generated_from_setfit_trainer
Model card Files Community

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:

  1. Fine-tuning a Sentence Transformer with contrastive learning.
  2. Training a classification head with features from the fine-tuned Sentence Transformer.

Model Details

Model Description

Model Sources

Model Labels

Label Examples
product faq
  • '8. What colors are available for the smart wallet?'
  • 'Is the Shoe Cleaning Brush - Combo of 3 a single product, kit, or combo?'
  • 'Can you describe the material of the Techbag Fingerlock Laptop Leather Bags in Brown?'
Out of Scope
  • 'I like to go swimming in the ocean'
  • 'The traffic was terrible on my way to work today'
  • 'Can I visit your physical store?'
order tracking
  • 'I need to know the expected delivery date for my order. Can you assist me with that?'
  • 'How long will it take to deliver the 50 pack of Brown Bakery Boxes to Patna?'
  • 'I need to know the status of my recent order. Can you check if it has been dispatched?'
general faq
  • 'How does hormonal imbalance contribute to acne?'
  • 'How to identify mashru silk'
  • 'What makes Purely Yours products different from other Ayurvedic brands?'
product policy
  • 'Are there any delivery charges for orders above INR 499?'
  • 'Do you offer a satisfaction guarantee for sneakers purchased with a store prepaid card?'
  • 'Do you share my personal information with third parties?'
product discoverability
  • 'What body products do you have for body mist?'
  • 'Are there any products available for female sexual wellness?'
  • 'Show me all bakery boxes with dividers'

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("What’s the best payment gateway for an online store?")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 4 10.4361 24
Label Training Sample Count
Out of Scope 60
general faq 60
order tracking 60
product discoverability 60
product faq 60
product policy 60

Training Hyperparameters

  • batch_size: (16, 16)
  • num_epochs: (2, 2)
  • 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: True

Training Results

Epoch Step Training Loss Validation Loss
0.0001 1 0.3001 -
0.0074 50 0.2362 -
0.0148 100 0.2573 -
0.0222 150 0.223 -
0.0296 200 0.2201 -
0.0370 250 0.1694 -
0.0444 300 0.0765 -
0.0519 350 0.0828 -
0.0593 400 0.0514 -
0.0667 450 0.0482 -
0.0741 500 0.0435 -
0.0815 550 0.006 -
0.0889 600 0.0071 -
0.0963 650 0.0049 -
0.1037 700 0.0023 -
0.1111 750 0.0007 -
0.1185 800 0.0011 -
0.1259 850 0.0007 -
0.1333 900 0.001 -
0.1407 950 0.0004 -
0.1481 1000 0.0006 -
0.1556 1050 0.001 -
0.1630 1100 0.0006 -
0.1704 1150 0.0006 -
0.1778 1200 0.0003 -
0.1852 1250 0.0003 -
0.1926 1300 0.0003 -
0.2 1350 0.0003 -
0.2074 1400 0.0003 -
0.2148 1450 0.0002 -
0.2222 1500 0.0004 -
0.2296 1550 0.0001 -
0.2370 1600 0.0002 -
0.2444 1650 0.0001 -
0.2519 1700 0.0002 -
0.2593 1750 0.0002 -
0.2667 1800 0.0002 -
0.2741 1850 0.0006 -
0.2815 1900 0.0024 -
0.2889 1950 0.0012 -
0.2963 2000 0.0002 -
0.3037 2050 0.0003 -
0.3111 2100 0.0002 -
0.3185 2150 0.0002 -
0.3259 2200 0.0001 -
0.3333 2250 0.0001 -
0.3407 2300 0.0001 -
0.3481 2350 0.0002 -
0.3556 2400 0.0001 -
0.3630 2450 0.0002 -
0.3704 2500 0.0001 -
0.3778 2550 0.0001 -
0.3852 2600 0.0019 -
0.3926 2650 0.0001 -
0.4 2700 0.0003 -
0.4074 2750 0.0001 -
0.4148 2800 0.0001 -
0.4222 2850 0.0001 -
0.4296 2900 0.0001 -
0.4370 2950 0.0002 -
0.4444 3000 0.0001 -
0.4519 3050 0.0001 -
0.4593 3100 0.0001 -
0.4667 3150 0.0001 -
0.4741 3200 0.0001 -
0.4815 3250 0.0001 -
0.4889 3300 0.0001 -
0.4963 3350 0.0001 -
0.5037 3400 0.0001 -
0.5111 3450 0.0001 -
0.5185 3500 0.0001 -
0.5259 3550 0.0001 -
0.5333 3600 0.0 -
0.5407 3650 0.0001 -
0.5481 3700 0.0001 -
0.5556 3750 0.0001 -
0.5630 3800 0.0001 -
0.5704 3850 0.0001 -
0.5778 3900 0.0 -
0.5852 3950 0.0001 -
0.5926 4000 0.0 -
0.6 4050 0.0001 -
0.6074 4100 0.0 -
0.6148 4150 0.0001 -
0.6222 4200 0.0001 -
0.6296 4250 0.0001 -
0.6370 4300 0.0001 -
0.6444 4350 0.0 -
0.6519 4400 0.0001 -
0.6593 4450 0.0001 -
0.6667 4500 0.0 -
0.6741 4550 0.0001 -
0.6815 4600 0.0001 -
0.6889 4650 0.0001 -
0.6963 4700 0.0001 -
0.7037 4750 0.0001 -
0.7111 4800 0.0001 -
0.7185 4850 0.0001 -
0.7259 4900 0.0001 -
0.7333 4950 0.0 -
0.7407 5000 0.0 -
0.7481 5050 0.0 -
0.7556 5100 0.0001 -
0.7630 5150 0.0001 -
0.7704 5200 0.0001 -
0.7778 5250 0.0001 -
0.7852 5300 0.0001 -
0.7926 5350 0.0 -
0.8 5400 0.0001 -
0.8074 5450 0.0 -
0.8148 5500 0.0001 -
0.8222 5550 0.0001 -
0.8296 5600 0.0 -
0.8370 5650 0.0 -
0.8444 5700 0.0 -
0.8519 5750 0.0001 -
0.8593 5800 0.0001 -
0.8667 5850 0.0001 -
0.8741 5900 0.0001 -
0.8815 5950 0.0001 -
0.8889 6000 0.0 -
0.8963 6050 0.0 -
0.9037 6100 0.0 -
0.9111 6150 0.0 -
0.9185 6200 0.0 -
0.9259 6250 0.0 -
0.9333 6300 0.0001 -
0.9407 6350 0.0 -
0.9481 6400 0.0 -
0.9556 6450 0.0 -
0.9630 6500 0.0 -
0.9704 6550 0.0 -
0.9778 6600 0.0 -
0.9852 6650 0.0001 -
0.9926 6700 0.0001 -
1.0 6750 0.0 -
1.0074 6800 0.0 -
1.0148 6850 0.0 -
1.0222 6900 0.0 -
1.0296 6950 0.0 -
1.0370 7000 0.0 -
1.0444 7050 0.0 -
1.0519 7100 0.0001 -
1.0593 7150 0.0 -
1.0667 7200 0.0 -
1.0741 7250 0.0003 -
1.0815 7300 0.0001 -
1.0889 7350 0.059 -
1.0963 7400 0.0002 -
1.1037 7450 0.0001 -
1.1111 7500 0.0001 -
1.1185 7550 0.0001 -
1.1259 7600 0.0001 -
1.1333 7650 0.0 -
1.1407 7700 0.0001 -
1.1481 7750 0.0 -
1.1556 7800 0.0001 -
1.1630 7850 0.0 -
1.1704 7900 0.0 -
1.1778 7950 0.0001 -
1.1852 8000 0.0001 -
1.1926 8050 0.0001 -
1.2 8100 0.0 -
1.2074 8150 0.0001 -
1.2148 8200 0.0001 -
1.2222 8250 0.0 -
1.2296 8300 0.0 -
1.2370 8350 0.0 -
1.2444 8400 0.0 -
1.2519 8450 0.0 -
1.2593 8500 0.0 -
1.2667 8550 0.0 -
1.2741 8600 0.0 -
1.2815 8650 0.0001 -
1.2889 8700 0.0 -
1.2963 8750 0.0 -
1.3037 8800 0.0 -
1.3111 8850 0.0 -
1.3185 8900 0.0 -
1.3259 8950 0.0 -
1.3333 9000 0.0 -
1.3407 9050 0.0 -
1.3481 9100 0.0 -
1.3556 9150 0.0 -
1.3630 9200 0.0 -
1.3704 9250 0.0 -
1.3778 9300 0.0 -
1.3852 9350 0.0 -
1.3926 9400 0.0 -
1.4 9450 0.0001 -
1.4074 9500 0.0 -
1.4148 9550 0.0 -
1.4222 9600 0.0 -
1.4296 9650 0.0 -
1.4370 9700 0.0 -
1.4444 9750 0.0 -
1.4519 9800 0.0 -
1.4593 9850 0.0 -
1.4667 9900 0.0 -
1.4741 9950 0.0 -
1.4815 10000 0.0 -
1.4889 10050 0.0 -
1.4963 10100 0.0 -
1.5037 10150 0.0 -
1.5111 10200 0.0 -
1.5185 10250 0.0 -
1.5259 10300 0.0 -
1.5333 10350 0.0 -
1.5407 10400 0.0 -
1.5481 10450 0.0 -
1.5556 10500 0.0 -
1.5630 10550 0.0 -
1.5704 10600 0.0 -
1.5778 10650 0.0 -
1.5852 10700 0.0 -
1.5926 10750 0.0 -
1.6 10800 0.0 -
1.6074 10850 0.0 -
1.6148 10900 0.0 -
1.6222 10950 0.0 -
1.6296 11000 0.0 -
1.6370 11050 0.0 -
1.6444 11100 0.0 -
1.6519 11150 0.0 -
1.6593 11200 0.0 -
1.6667 11250 0.0 -
1.6741 11300 0.0 -
1.6815 11350 0.0 -
1.6889 11400 0.0 -
1.6963 11450 0.0 -
1.7037 11500 0.0 -
1.7111 11550 0.0 -
1.7185 11600 0.0 -
1.7259 11650 0.0001 -
1.7333 11700 0.0 -
1.7407 11750 0.0 -
1.7481 11800 0.0 -
1.7556 11850 0.0 -
1.7630 11900 0.0 -
1.7704 11950 0.0 -
1.7778 12000 0.0 -
1.7852 12050 0.0 -
1.7926 12100 0.0 -
1.8 12150 0.0 -
1.8074 12200 0.0 -
1.8148 12250 0.0 -
1.8222 12300 0.0 -
1.8296 12350 0.0 -
1.8370 12400 0.0 -
1.8444 12450 0.0 -
1.8519 12500 0.0 -
1.8593 12550 0.0 -
1.8667 12600 0.0 -
1.8741 12650 0.0 -
1.8815 12700 0.0 -
1.8889 12750 0.0 -
1.8963 12800 0.0 -
1.9037 12850 0.0 -
1.9111 12900 0.0 -
1.9185 12950 0.0 -
1.9259 13000 0.0 -
1.9333 13050 0.0 -
1.9407 13100 0.0 -
1.9481 13150 0.0 -
1.9556 13200 0.0 -
1.9630 13250 0.0 -
1.9704 13300 0.0 -
1.9778 13350 0.0 -
1.9852 13400 0.0 -
1.9926 13450 0.0 -
2.0 13500 0.0 -

Framework Versions

  • Python: 3.10.16
  • SetFit: 1.0.3
  • Sentence Transformers: 2.7.0
  • Transformers: 4.40.2
  • PyTorch: 2.2.2
  • Datasets: 2.19.1
  • 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}
}
Downloads last month
1
Safetensors
Model size
109M params
Tensor type
F32
·
Inference Providers NEW
Text Classification
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Shankhdhar/classifier_woog_arista_vault_combined

Base model

sentence-transformers/paraphrase-mpnet-base-v2
Finetuned
(295)
this model