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
non-conspiratorial	'Vaccines are safe, effective, and prevent serious diseases.' "The sun's distance from Earth is well-documented by astronomers." 'The US government investigates and responds to threats like terrorism.'
conspiratorial	'The music industry is controlled by occultists.' 'The assassination of Abraham Lincoln was a larger conspiracy.' 'Freemasons are involved in a global conspiracy.'

Evaluation

Metrics

Label	Accuracy
all	1.0

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("annarae/setfit_model_30Epoch_17Aug")
# Run inference
preds = model("Food safety is regulated to protect public health.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	5	8.8438	13

Label	Training Sample Count
conspiratorial	156
non-conspiratorial	164

Training Hyperparameters

batch_size: (16, 16)
num_epochs: (30, 30)
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
0.0013	1	0.3267	-
0.0625	50	0.3543	-
0.125	100	0.3361	-
0.1875	150	0.2603	-
0.25	200	0.142	-
0.3125	250	0.0563	-
0.375	300	0.0337	-
0.4375	350	0.0036	-
0.5	400	0.0024	-
0.5625	450	0.0011	-
0.625	500	0.0015	-
0.6875	550	0.0007	-
0.75	600	0.0006	-
0.8125	650	0.0004	-
0.875	700	0.0004	-
0.9375	750	0.0004	-
1.0	800	0.0004	-
1.0625	850	0.0003	-
1.125	900	0.0003	-
1.1875	950	0.0003	-
1.25	1000	0.0002	-
1.3125	1050	0.0002	-
1.375	1100	0.0003	-
1.4375	1150	0.0002	-
1.5	1200	0.0002	-
1.5625	1250	0.0002	-
1.625	1300	0.0002	-
1.6875	1350	0.0001	-
1.75	1400	0.0001	-
1.8125	1450	0.0001	-
1.875	1500	0.0001	-
1.9375	1550	0.0001	-
2.0	1600	0.0001	-
2.0625	1650	0.0001	-
2.125	1700	0.0001	-
2.1875	1750	0.0001	-
2.25	1800	0.0001	-
2.3125	1850	0.0001	-
2.375	1900	0.0001	-
2.4375	1950	0.0001	-
2.5	2000	0.0001	-
2.5625	2050	0.0001	-
2.625	2100	0.0001	-
2.6875	2150	0.0001	-
2.75	2200	0.0001	-
2.8125	2250	0.0001	-
2.875	2300	0.0001	-
2.9375	2350	0.0001	-
3.0	2400	0.0001	-
3.0625	2450	0.0001	-
3.125	2500	0.0001	-
3.1875	2550	0.0001	-
3.25	2600	0.0009	-
3.3125	2650	0.0012	-
3.375	2700	0.0001	-
3.4375	2750	0.0001	-
3.5	2800	0.0001	-
3.5625	2850	0.0001	-
3.625	2900	0.0001	-
3.6875	2950	0.0001	-
3.75	3000	0.0001	-
3.8125	3050	0.0001	-
3.875	3100	0.0	-
3.9375	3150	0.0001	-
4.0	3200	0.0	-
4.0625	3250	0.0	-
4.125	3300	0.0	-
4.1875	3350	0.0	-
4.25	3400	0.0	-
4.3125	3450	0.0	-
4.375	3500	0.0	-
4.4375	3550	0.0	-
4.5	3600	0.0	-
4.5625	3650	0.0	-
4.625	3700	0.0	-
4.6875	3750	0.0	-
4.75	3800	0.0	-
4.8125	3850	0.0	-
4.875	3900	0.0	-
4.9375	3950	0.0	-
5.0	4000	0.0	-
5.0625	4050	0.0	-
5.125	4100	0.0	-
5.1875	4150	0.0	-
5.25	4200	0.0	-
5.3125	4250	0.0	-
5.375	4300	0.0	-
5.4375	4350	0.0	-
5.5	4400	0.0	-
5.5625	4450	0.0	-
5.625	4500	0.0	-
5.6875	4550	0.0	-
5.75	4600	0.0	-
5.8125	4650	0.0	-
5.875	4700	0.0	-
5.9375	4750	0.0	-
6.0	4800	0.0	-
6.0625	4850	0.0	-
6.125	4900	0.0	-
6.1875	4950	0.0	-
6.25	5000	0.0	-
6.3125	5050	0.0	-
6.375	5100	0.0	-
6.4375	5150	0.0	-
6.5	5200	0.0	-
6.5625	5250	0.0	-
6.625	5300	0.0	-
6.6875	5350	0.0	-
6.75	5400	0.0	-
6.8125	5450	0.0	-
6.875	5500	0.0	-
6.9375	5550	0.0	-
7.0	5600	0.0	-
7.0625	5650	0.0	-
7.125	5700	0.0	-
7.1875	5750	0.0	-
7.25	5800	0.0	-
7.3125	5850	0.0	-
7.375	5900	0.0	-
7.4375	5950	0.0	-
7.5	6000	0.0	-
7.5625	6050	0.0	-
7.625	6100	0.0	-
7.6875	6150	0.0	-
7.75	6200	0.0	-
7.8125	6250	0.0	-
7.875	6300	0.0	-
7.9375	6350	0.0	-
8.0	6400	0.0	-
8.0625	6450	0.0	-
8.125	6500	0.0	-
8.1875	6550	0.0	-
8.25	6600	0.0	-
8.3125	6650	0.0	-
8.375	6700	0.0	-
8.4375	6750	0.0	-
8.5	6800	0.0	-
8.5625	6850	0.0	-
8.625	6900	0.0	-
8.6875	6950	0.0	-
8.75	7000	0.0	-
8.8125	7050	0.0	-
8.875	7100	0.0	-
8.9375	7150	0.0	-
9.0	7200	0.0	-
9.0625	7250	0.0	-
9.125	7300	0.0	-
9.1875	7350	0.0	-
9.25	7400	0.0	-
9.3125	7450	0.0	-
9.375	7500	0.0	-
9.4375	7550	0.0	-
9.5	7600	0.0	-
9.5625	7650	0.0	-
9.625	7700	0.0	-
9.6875	7750	0.0	-
9.75	7800	0.0	-
9.8125	7850	0.0	-
9.875	7900	0.0	-
9.9375	7950	0.0	-
10.0	8000	0.0	-
10.0625	8050	0.0	-
10.125	8100	0.0	-
10.1875	8150	0.0	-
10.25	8200	0.0	-
10.3125	8250	0.0	-
10.375	8300	0.0	-
10.4375	8350	0.0	-
10.5	8400	0.0	-
10.5625	8450	0.0	-
10.625	8500	0.0	-
10.6875	8550	0.0	-
10.75	8600	0.0	-
10.8125	8650	0.0	-
10.875	8700	0.0	-
10.9375	8750	0.0	-
11.0	8800	0.0	-
11.0625	8850	0.0	-
11.125	8900	0.0	-
11.1875	8950	0.0	-
11.25	9000	0.0	-
11.3125	9050	0.0	-
11.375	9100	0.0	-
11.4375	9150	0.0	-
11.5	9200	0.0	-
11.5625	9250	0.0	-
11.625	9300	0.0	-
11.6875	9350	0.0	-
11.75	9400	0.0	-
11.8125	9450	0.0	-
11.875	9500	0.0	-
11.9375	9550	0.0	-
12.0	9600	0.0	-
12.0625	9650	0.0	-
12.125	9700	0.0	-
12.1875	9750	0.0	-
12.25	9800	0.0	-
12.3125	9850	0.0	-
12.375	9900	0.0	-
12.4375	9950	0.0	-
12.5	10000	0.0	-
12.5625	10050	0.0	-
12.625	10100	0.0	-
12.6875	10150	0.0	-
12.75	10200	0.0	-
12.8125	10250	0.0	-
12.875	10300	0.0	-
12.9375	10350	0.0	-
13.0	10400	0.0	-
13.0625	10450	0.0	-
13.125	10500	0.0	-
13.1875	10550	0.0	-
13.25	10600	0.0	-
13.3125	10650	0.0	-
13.375	10700	0.0	-
13.4375	10750	0.0	-
13.5	10800	0.0	-
13.5625	10850	0.0	-
13.625	10900	0.0	-
13.6875	10950	0.0	-
13.75	11000	0.0	-
13.8125	11050	0.0	-
13.875	11100	0.0	-
13.9375	11150	0.0	-
14.0	11200	0.0	-
14.0625	11250	0.0	-
14.125	11300	0.0	-
14.1875	11350	0.0	-
14.25	11400	0.0	-
14.3125	11450	0.0	-
14.375	11500	0.0	-
14.4375	11550	0.0	-
14.5	11600	0.0	-
14.5625	11650	0.0	-
14.625	11700	0.0	-
14.6875	11750	0.0	-
14.75	11800	0.0	-
14.8125	11850	0.0	-
14.875	11900	0.0	-
14.9375	11950	0.0	-
15.0	12000	0.0	-
15.0625	12050	0.0	-
15.125	12100	0.0	-
15.1875	12150	0.0	-
15.25	12200	0.0	-
15.3125	12250	0.0	-
15.375	12300	0.0	-
15.4375	12350	0.0	-
15.5	12400	0.0	-
15.5625	12450	0.0	-
15.625	12500	0.0	-
15.6875	12550	0.0	-
15.75	12600	0.0	-
15.8125	12650	0.0	-
15.875	12700	0.0	-
15.9375	12750	0.0	-
16.0	12800	0.0	-
16.0625	12850	0.0	-
16.125	12900	0.0	-
16.1875	12950	0.0	-
16.25	13000	0.0	-
16.3125	13050	0.0	-
16.375	13100	0.0	-
16.4375	13150	0.0	-
16.5	13200	0.0	-
16.5625	13250	0.0	-
16.625	13300	0.0	-
16.6875	13350	0.0	-
16.75	13400	0.0	-
16.8125	13450	0.0	-
16.875	13500	0.0	-
16.9375	13550	0.0	-
17.0	13600	0.0	-
17.0625	13650	0.0	-
17.125	13700	0.0	-
17.1875	13750	0.0	-
17.25	13800	0.0	-
17.3125	13850	0.0	-
17.375	13900	0.0	-
17.4375	13950	0.0	-
17.5	14000	0.0	-
17.5625	14050	0.0	-
17.625	14100	0.0	-
17.6875	14150	0.0	-
17.75	14200	0.0	-
17.8125	14250	0.0	-
17.875	14300	0.0	-
17.9375	14350	0.0	-
18.0	14400	0.0	-
18.0625	14450	0.0	-
18.125	14500	0.0	-
18.1875	14550	0.0	-
18.25	14600	0.0	-
18.3125	14650	0.0	-
18.375	14700	0.0	-
18.4375	14750	0.0	-
18.5	14800	0.0	-
18.5625	14850	0.0	-
18.625	14900	0.0	-
18.6875	14950	0.0	-
18.75	15000	0.0	-
18.8125	15050	0.0	-
18.875	15100	0.0	-
18.9375	15150	0.0	-
19.0	15200	0.0	-
19.0625	15250	0.0	-
19.125	15300	0.0	-
19.1875	15350	0.0	-
19.25	15400	0.0	-
19.3125	15450	0.0	-
19.375	15500	0.0	-
19.4375	15550	0.0	-
19.5	15600	0.0	-
19.5625	15650	0.0	-
19.625	15700	0.0	-
19.6875	15750	0.0	-
19.75	15800	0.0	-
19.8125	15850	0.0	-
19.875	15900	0.0	-
19.9375	15950	0.0	-
20.0	16000	0.0	-
20.0625	16050	0.0	-
20.125	16100	0.0	-
20.1875	16150	0.0	-
20.25	16200	0.0	-
20.3125	16250	0.0	-
20.375	16300	0.0	-
20.4375	16350	0.0	-
20.5	16400	0.0	-
20.5625	16450	0.0	-
20.625	16500	0.0	-
20.6875	16550	0.0	-
20.75	16600	0.0	-
20.8125	16650	0.0	-
20.875	16700	0.0	-
20.9375	16750	0.0	-
21.0	16800	0.0	-
21.0625	16850	0.0	-
21.125	16900	0.0	-
21.1875	16950	0.0	-
21.25	17000	0.0	-
21.3125	17050	0.0	-
21.375	17100	0.0	-
21.4375	17150	0.0	-
21.5	17200	0.0	-
21.5625	17250	0.0	-
21.625	17300	0.0	-
21.6875	17350	0.0	-
21.75	17400	0.0	-
21.8125	17450	0.0	-
21.875	17500	0.0	-
21.9375	17550	0.0	-
22.0	17600	0.0	-
22.0625	17650	0.0	-
22.125	17700	0.0	-
22.1875	17750	0.0	-
22.25	17800	0.0	-
22.3125	17850	0.0	-
22.375	17900	0.0	-
22.4375	17950	0.0	-
22.5	18000	0.0	-
22.5625	18050	0.0	-
22.625	18100	0.0	-
22.6875	18150	0.0	-
22.75	18200	0.0	-
22.8125	18250	0.0	-
22.875	18300	0.0	-
22.9375	18350	0.0	-
23.0	18400	0.0	-
23.0625	18450	0.0	-
23.125	18500	0.0	-
23.1875	18550	0.0	-
23.25	18600	0.0	-
23.3125	18650	0.0	-
23.375	18700	0.0	-
23.4375	18750	0.0	-
23.5	18800	0.0	-
23.5625	18850	0.0	-
23.625	18900	0.0	-
23.6875	18950	0.0	-
23.75	19000	0.0	-
23.8125	19050	0.0	-
23.875	19100	0.0	-
23.9375	19150	0.0	-
24.0	19200	0.0	-
24.0625	19250	0.0	-
24.125	19300	0.0	-
24.1875	19350	0.0	-
24.25	19400	0.0	-
24.3125	19450	0.0	-
24.375	19500	0.0	-
24.4375	19550	0.0	-
24.5	19600	0.0	-
24.5625	19650	0.0	-
24.625	19700	0.0	-
24.6875	19750	0.0	-
24.75	19800	0.0	-
24.8125	19850	0.0	-
24.875	19900	0.0	-
24.9375	19950	0.0	-
25.0	20000	0.0	-
25.0625	20050	0.0	-
25.125	20100	0.0	-
25.1875	20150	0.0	-
25.25	20200	0.0	-
25.3125	20250	0.0	-
25.375	20300	0.0	-
25.4375	20350	0.0	-
25.5	20400	0.0	-
25.5625	20450	0.0	-
25.625	20500	0.0	-
25.6875	20550	0.0	-
25.75	20600	0.0	-
25.8125	20650	0.0	-
25.875	20700	0.0	-
25.9375	20750	0.0	-
26.0	20800	0.0	-
26.0625	20850	0.0	-
26.125	20900	0.0	-
26.1875	20950	0.0	-
26.25	21000	0.0	-
26.3125	21050	0.0	-
26.375	21100	0.0	-
26.4375	21150	0.0	-
26.5	21200	0.0	-
26.5625	21250	0.0	-
26.625	21300	0.0	-
26.6875	21350	0.0	-
26.75	21400	0.0	-
26.8125	21450	0.0	-
26.875	21500	0.0	-
26.9375	21550	0.0	-
27.0	21600	0.0	-
27.0625	21650	0.0	-
27.125	21700	0.0	-
27.1875	21750	0.0	-
27.25	21800	0.0	-
27.3125	21850	0.0	-
27.375	21900	0.0	-
27.4375	21950	0.0	-
27.5	22000	0.0	-
27.5625	22050	0.0	-
27.625	22100	0.0	-
27.6875	22150	0.0	-
27.75	22200	0.0	-
27.8125	22250	0.0	-
27.875	22300	0.0	-
27.9375	22350	0.0	-
28.0	22400	0.0	-
28.0625	22450	0.0	-
28.125	22500	0.0	-
28.1875	22550	0.0	-
28.25	22600	0.0	-
28.3125	22650	0.0	-
28.375	22700	0.0	-
28.4375	22750	0.0	-
28.5	22800	0.0	-
28.5625	22850	0.0	-
28.625	22900	0.0	-
28.6875	22950	0.0	-
28.75	23000	0.0	-
28.8125	23050	0.0	-
28.875	23100	0.0	-
28.9375	23150	0.0	-
29.0	23200	0.0	-
29.0625	23250	0.0	-
29.125	23300	0.0	-
29.1875	23350	0.0	-
29.25	23400	0.0	-
29.3125	23450	0.0	-
29.375	23500	0.0	-
29.4375	23550	0.0	-
29.5	23600	0.0	-
29.5625	23650	0.0	-
29.625	23700	0.0	-
29.6875	23750	0.0	-
29.75	23800	0.0	-
29.8125	23850	0.0	-
29.875	23900	0.0	-
29.9375	23950	0.0	-
30.0	24000	0.0	-

Framework Versions

Python: 3.10.12
SetFit: 1.0.3
Sentence Transformers: 3.0.1
Transformers: 4.39.0
PyTorch: 2.3.1+cu121
Datasets: 2.21.0
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}
}

annarae
/

setfit_model_30Epoch_17Aug

You need to agree to share your contact information to access this model