Hugging Face's logo

richie-ghost
/
setfit-FacebookAI-roberta-base-phatic

Text Classification
setfit
Safetensors
sentence-transformers
roberta
generated_from_setfit_trainer
Eval Results
Model card Files Community
Edit model card

SetFit with FacebookAI/roberta-base

This is a SetFit model that can be used for Text Classification. This SetFit model uses FacebookAI/roberta-base 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
true
  • 'How can we apply your findings to optimize our processes?'
  • 'Your presence at the meeting was greatly appreciated.'
  • 'Your journey is quite inspiring, can you share more about it?'
false
  • 'What book are you currently reading?'
  • 'It’s important to acknowledge your feelings, what’s been going through your mind?'
  • 'You’ve been working hard on your mental health; how are you finding the journey?'

Evaluation

Metrics

Label Accuracy
all 0.94

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("richie-ghost/setfit-FacebookAI-roberta-base-phatic")
# Run inference
preds = model("Take it easy!")

Training Details

Training Set Metrics

Training set Min Median Max
Word count 1 9.8722 108
Label Training Sample Count
false 191
true 169

Training Hyperparameters

  • batch_size: (16, 16)
  • num_epochs: (4, 4)
  • 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
0.0002 1 0.4475 -
0.0122 50 0.4363 -
0.0245 100 0.3668 -
0.0367 150 0.177 -
0.0489 200 0.0999 -
0.0612 250 0.1043 -
0.0734 300 0.0191 -
0.0856 350 0.009 -
0.0978 400 0.0028 -
0.1101 450 0.0046 -
0.1223 500 0.0012 -
0.1345 550 0.0016 -
0.1468 600 0.0012 -
0.1590 650 0.0012 -
0.1712 700 0.0164 -
0.1835 750 0.025 -
0.1957 800 0.0007 -
0.2079 850 0.0013 -
0.2202 900 0.0008 -
0.2324 950 0.0005 -
0.2446 1000 0.0004 -
0.2568 1050 0.0002 -
0.2691 1100 0.0004 -
0.2813 1150 0.0003 -
0.2935 1200 0.0002 -
0.3058 1250 0.0002 -
0.3180 1300 0.0003 -
0.3302 1350 0.0002 -
0.3425 1400 0.0001 -
0.3547 1450 0.003 -
0.3669 1500 0.0003 -
0.3792 1550 0.0003 -
0.3914 1600 0.0001 -
0.4036 1650 0.0001 -
0.4159 1700 0.0001 -
0.4281 1750 0.0001 -
0.4403 1800 0.0001 -
0.4525 1850 0.0001 -
0.4648 1900 0.0001 -
0.4770 1950 0.0001 -
0.4892 2000 0.0001 -
0.5015 2050 0.0001 -
0.5137 2100 0.0001 -
0.5259 2150 0.0001 -
0.5382 2200 0.0 -
0.5504 2250 0.0 -
0.5626 2300 0.0 -
0.5749 2350 0.0001 -
0.5871 2400 0.0 -
0.5993 2450 0.0 -
0.6115 2500 0.0001 -
0.6238 2550 0.0001 -
0.6360 2600 0.0 -
0.6482 2650 0.0 -
0.6605 2700 0.0 -
0.6727 2750 0.0 -
0.6849 2800 0.0 -
0.6972 2850 0.0 -
0.7094 2900 0.0001 -
0.7216 2950 0.0001 -
0.7339 3000 0.0 -
0.7461 3050 0.0 -
0.7583 3100 0.0006 -
0.7705 3150 0.0606 -
0.7828 3200 0.0 -
0.7950 3250 0.0002 -
0.8072 3300 0.0 -
0.8195 3350 0.0001 -
0.8317 3400 0.0001 -
0.8439 3450 0.0 -
0.8562 3500 0.0001 -
0.8684 3550 0.0 -
0.8806 3600 0.0 -
0.8929 3650 0.0 -
0.9051 3700 0.0 -
0.9173 3750 0.0 -
0.9295 3800 0.0 -
0.9418 3850 0.0 -
0.9540 3900 0.0 -
0.9662 3950 0.0 -
0.9785 4000 0.0 -
0.9907 4050 0.0 -
1.0 4088 - 0.1621
1.0029 4100 0.0 -
1.0152 4150 0.0 -
1.0274 4200 0.0 -
1.0396 4250 0.0 -
1.0519 4300 0.0 -
1.0641 4350 0.0 -
1.0763 4400 0.0 -
1.0886 4450 0.0 -
1.1008 4500 0.0 -
1.1130 4550 0.0 -
1.1252 4600 0.0 -
1.1375 4650 0.0 -
1.1497 4700 0.0 -
1.1619 4750 0.0 -
1.1742 4800 0.0 -
1.1864 4850 0.0 -
1.1986 4900 0.0 -
1.2109 4950 0.0 -
1.2231 5000 0.0 -
1.2353 5050 0.0 -
1.2476 5100 0.0 -
1.2598 5150 0.0 -
1.2720 5200 0.0 -
1.2842 5250 0.0 -
1.2965 5300 0.0 -
1.3087 5350 0.0 -
1.3209 5400 0.0 -
1.3332 5450 0.0 -
1.3454 5500 0.0 -
1.3576 5550 0.0 -
1.3699 5600 0.0 -
1.3821 5650 0.0 -
1.3943 5700 0.0 -
1.4066 5750 0.0 -
1.4188 5800 0.0 -
1.4310 5850 0.0 -
1.4432 5900 0.0 -
1.4555 5950 0.0 -
1.4677 6000 0.0 -
1.4799 6050 0.0 -
1.4922 6100 0.0 -
1.5044 6150 0.0 -
1.5166 6200 0.0 -
1.5289 6250 0.0 -
1.5411 6300 0.0 -
1.5533 6350 0.0 -
1.5656 6400 0.0 -
1.5778 6450 0.0 -
1.5900 6500 0.0 -
1.6023 6550 0.0 -
1.6145 6600 0.0 -
1.6267 6650 0.0 -
1.6389 6700 0.0 -
1.6512 6750 0.0 -
1.6634 6800 0.0 -
1.6756 6850 0.0 -
1.6879 6900 0.0 -
1.7001 6950 0.0 -
1.7123 7000 0.0 -
1.7246 7050 0.0 -
1.7368 7100 0.0 -
1.7490 7150 0.0 -
1.7613 7200 0.0 -
1.7735 7250 0.0 -
1.7857 7300 0.0 -
1.7979 7350 0.0 -
1.8102 7400 0.0 -
1.8224 7450 0.0 -
1.8346 7500 0.0 -
1.8469 7550 0.0 -
1.8591 7600 0.0 -
1.8713 7650 0.0 -
1.8836 7700 0.0 -
1.8958 7750 0.0 -
1.9080 7800 0.0 -
1.9203 7850 0.0 -
1.9325 7900 0.0 -
1.9447 7950 0.0 -
1.9569 8000 0.0 -
1.9692 8050 0.0 -
1.9814 8100 0.0 -
1.9936 8150 0.0 -
2.0 8176 - 0.1131
2.0059 8200 0.0 -
2.0181 8250 0.0 -
2.0303 8300 0.0 -
2.0426 8350 0.0 -
2.0548 8400 0.0 -
2.0670 8450 0.0 -
2.0793 8500 0.0 -
2.0915 8550 0.0 -
2.1037 8600 0.0 -
2.1159 8650 0.0 -
2.1282 8700 0.0 -
2.1404 8750 0.0 -
2.1526 8800 0.0 -
2.1649 8850 0.0 -
2.1771 8900 0.0 -
2.1893 8950 0.0 -
2.2016 9000 0.0 -
2.2138 9050 0.0 -
2.2260 9100 0.0 -
2.2383 9150 0.0 -
2.2505 9200 0.0 -
2.2627 9250 0.0 -
2.2750 9300 0.0 -
2.2872 9350 0.0 -
2.2994 9400 0.0 -
2.3116 9450 0.0 -
2.3239 9500 0.0 -
2.3361 9550 0.0 -
2.3483 9600 0.0 -
2.3606 9650 0.0 -
2.3728 9700 0.0 -
2.3850 9750 0.0 -
2.3973 9800 0.0 -
2.4095 9850 0.0 -
2.4217 9900 0.0 -
2.4340 9950 0.0 -
2.4462 10000 0.0 -
2.4584 10050 0.0 -
2.4706 10100 0.0 -
2.4829 10150 0.0 -
2.4951 10200 0.0 -
2.5073 10250 0.0 -
2.5196 10300 0.0 -
2.5318 10350 0.0 -
2.5440 10400 0.0 -
2.5563 10450 0.0 -
2.5685 10500 0.0 -
2.5807 10550 0.0 -
2.5930 10600 0.0 -
2.6052 10650 0.0 -
2.6174 10700 0.0 -
2.6296 10750 0.0 -
2.6419 10800 0.0 -
2.6541 10850 0.0 -
2.6663 10900 0.0 -
2.6786 10950 0.0 -
2.6908 11000 0.0 -
2.7030 11050 0.0 -
2.7153 11100 0.0 -
2.7275 11150 0.0 -
2.7397 11200 0.0 -
2.7520 11250 0.0 -
2.7642 11300 0.0 -
2.7764 11350 0.0 -
2.7886 11400 0.0 -
2.8009 11450 0.0 -
2.8131 11500 0.0 -
2.8253 11550 0.0 -
2.8376 11600 0.0 -
2.8498 11650 0.0 -
2.8620 11700 0.0 -
2.8743 11750 0.0 -
2.8865 11800 0.0 -
2.8987 11850 0.0 -
2.9110 11900 0.0 -
2.9232 11950 0.0 -
2.9354 12000 0.0 -
2.9477 12050 0.0 -
2.9599 12100 0.0 -
2.9721 12150 0.0 -
2.9843 12200 0.0 -
2.9966 12250 0.0 -
3.0 12264 - 0.1127
3.0088 12300 0.0 -
3.0210 12350 0.0 -
3.0333 12400 0.0 -
3.0455 12450 0.0 -
3.0577 12500 0.0 -
3.0700 12550 0.0 -
3.0822 12600 0.0 -
3.0944 12650 0.0 -
3.1067 12700 0.0 -
3.1189 12750 0.0 -
3.1311 12800 0.0 -
3.1433 12850 0.0 -
3.1556 12900 0.0 -
3.1678 12950 0.0 -
3.1800 13000 0.0 -
3.1923 13050 0.0 -
3.2045 13100 0.0 -
3.2167 13150 0.0 -
3.2290 13200 0.0 -
3.2412 13250 0.0 -
3.2534 13300 0.0 -
3.2657 13350 0.0 -
3.2779 13400 0.0 -
3.2901 13450 0.0 -
3.3023 13500 0.0 -
3.3146 13550 0.0 -
3.3268 13600 0.0 -
3.3390 13650 0.0 -
3.3513 13700 0.0 -
3.3635 13750 0.0 -
3.3757 13800 0.0 -
3.3880 13850 0.0 -
3.4002 13900 0.0 -
3.4124 13950 0.0 -
3.4247 14000 0.0 -
3.4369 14050 0.0 -
3.4491 14100 0.0 -
3.4614 14150 0.0 -
3.4736 14200 0.0 -
3.4858 14250 0.0 -
3.4980 14300 0.0 -
3.5103 14350 0.0 -
3.5225 14400 0.0 -
3.5347 14450 0.0 -
3.5470 14500 0.0 -
3.5592 14550 0.0 -
3.5714 14600 0.0 -
3.5837 14650 0.0 -
3.5959 14700 0.0 -
3.6081 14750 0.0 -
3.6204 14800 0.0 -
3.6326 14850 0.0 -
3.6448 14900 0.0 -
3.6570 14950 0.0 -
3.6693 15000 0.0 -
3.6815 15050 0.0 -
3.6937 15100 0.0 -
3.7060 15150 0.0 -
3.7182 15200 0.0 -
3.7304 15250 0.0 -
3.7427 15300 0.0 -
3.7549 15350 0.0 -
3.7671 15400 0.0 -
3.7794 15450 0.0 -
3.7916 15500 0.0 -
3.8038 15550 0.0 -
3.8160 15600 0.0 -
3.8283 15650 0.0 -
3.8405 15700 0.0 -
3.8527 15750 0.0 -
3.8650 15800 0.0 -
3.8772 15850 0.0 -
3.8894 15900 0.0 -
3.9017 15950 0.0 -
3.9139 16000 0.0 -
3.9261 16050 0.0 -
3.9384 16100 0.0 -
3.9506 16150 0.0 -
3.9628 16200 0.0 -
3.9750 16250 0.0 -
3.9873 16300 0.0 -
3.9995 16350 0.0 -
4.0 16352 - 0.1019
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.12
  • SetFit: 1.0.3
  • Sentence Transformers: 2.7.0
  • Transformers: 4.40.0
  • 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}
}
Downloads last month
8
Safetensors
Model size
125M params
Tensor type
F32
·

Finetuned from

Evaluation results

View on Papers With Code