bhaskars113/toyota-paint-attribute-1.1

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

This is a SetFit model trained on the bhaskars113/toyota-paint-attributes dataset that can be used for Text Classification. This SetFit model uses sentence-transformers/all-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/all-mpnet-base-v2
• Classification head: a OneVsRestClassifier instance
• Maximum Sequence Length: 384 tokens
• Training Dataset: bhaskars113/toyota-paint-attributes

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/toyota-paint-attribute-1.1")
# Run inference
preds = model("The car is from Utah and garage kept, so the paint is still in very good condition")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	5	33.8098	155

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	0.1664	-
0.0196	50	0.2377	-
0.0392	100	0.1178	-
0.0588	150	0.0577	-
0.0784	200	0.0163	-
0.0980	250	0.0265	-
0.1176	300	0.0867	-
0.1373	350	0.0181	-
0.1569	400	0.0153	-
0.1765	450	0.0411	-
0.1961	500	0.0308	-
0.2157	550	0.0258	-
0.2353	600	0.0062	-
0.2549	650	0.0036	-
0.2745	700	0.0087	-
0.2941	750	0.0025	-
0.3137	800	0.004	-
0.3333	850	0.0025	-
0.3529	900	0.0044	-
0.3725	950	0.0031	-
0.3922	1000	0.0018	-
0.4118	1050	0.0046	-
0.4314	1100	0.0013	-
0.4510	1150	0.0014	-
0.4706	1200	0.002	-
0.4902	1250	0.0015	-
0.5098	1300	0.0039	-
0.5294	1350	0.0019	-
0.5490	1400	0.0011	-
0.5686	1450	0.0008	-
0.5882	1500	0.0015	-
0.6078	1550	0.0012	-
0.6275	1600	0.0011	-
0.6471	1650	0.0008	-
0.6667	1700	0.0016	-
0.6863	1750	0.0009	-
0.7059	1800	0.0008	-
0.7255	1850	0.0008	-
0.7451	1900	0.0008	-
0.7647	1950	0.0011	-
0.7843	2000	0.0008	-
0.8039	2050	0.001	-
0.8235	2100	0.001	-
0.8431	2150	0.0009	-
0.8627	2200	0.0067	-
0.8824	2250	0.0008	-
0.9020	2300	0.0009	-
0.9216	2350	0.0009	-
0.9412	2400	0.0007	-
0.9608	2450	0.0006	-
0.9804	2500	0.0007	-
1.0	2550	0.0006	-

Framework Versions

• Python: 3.10.12
• SetFit: 1.0.3
• Sentence Transformers: 2.7.0
• Transformers: 4.40.2
• PyTorch: 2.2.1+cu121
• 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}
}