bhaskars113/go-emotions-multilabel

SetFit with sentence-transformers/paraphrase-mpnet-base-v2

This is a SetFit model trained on the go_emotions dataset that can be used for Text Classification. This SetFit model uses sentence-transformers/paraphrase-mpnet-base-v2 as the Sentence Transformer embedding model. A OneVsRestClassifier 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 Type: SetFit
• Sentence Transformer body: sentence-transformers/paraphrase-mpnet-base-v2
• Classification head: a OneVsRestClassifier instance
• Maximum Sequence Length: 512 tokens
• Training Dataset: go_emotions

Model Sources

• Repository: SetFit on GitHub
• Paper: Efficient Few-Shot Learning Without Prompts
• Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

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("bhaskars113/go-emotions-multilabel")
# Run inference
preds = model("I think you mean the announcement")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	1	13.6060	30

Training Hyperparameters

• batch_size: (16, 16)
• num_epochs: (1, 1)
• 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
0.0004	1	0.3873	-
0.0223	50	0.2243	-
0.0446	100	0.2305	-
0.0670	150	0.2297	-
0.0893	200	0.2758	-
0.1116	250	0.2197	-
0.1339	300	0.1984	-
0.1562	350	0.1729	-
0.1786	400	0.1244	-
0.2009	450	0.164	-
0.2232	500	0.1587	-
0.2455	550	0.2272	-
0.2679	600	0.3367	-
0.2902	650	0.1715	-
0.3125	700	0.2213	-
0.3348	750	0.2394	-
0.3571	800	0.1275	-
0.3795	850	0.1919	-
0.4018	900	0.143	-
0.4241	950	0.2431	-
0.4464	1000	0.1747	-
0.4688	1050	0.1567	-
0.4911	1100	0.194	-
0.5134	1150	0.1895	-
0.5357	1200	0.1601	-
0.5580	1250	0.1042	-
0.5804	1300	0.0553	-
0.6027	1350	0.1614	-
0.625	1400	0.1854	-
0.6473	1450	0.1259	-
0.6696	1500	0.138	-
0.6920	1550	0.2181	-
0.7143	1600	0.1144	-
0.7366	1650	0.1987	-
0.7589	1700	0.0859	-
0.7812	1750	0.1665	-
0.8036	1800	0.1628	-
0.8259	1850	0.2296	-
0.8482	1900	0.1892	-
0.8705	1950	0.2033	-
0.8929	2000	0.1507	-
0.9152	2050	0.1592	-
0.9375	2100	0.1077	-
0.9598	2150	0.1415	-
0.9821	2200	0.1561	-

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}
}