--- tags: - setfit - sentence-transformers - text-classification - generated_from_setfit_trainer widget: - text: The infrastructure requirement for collection based on the targets and projections made is presented in Table 12.22. A total of about 149,000 km length and 8,660km length of sewers are required for urban and rural communities, respectively by 2047. In addition, a little over 4 million facilities in urban areas and about 853,000 facilities in rural areas will be required to meet on-site sanitation needs by 2033 nationwide. - text: The population of the Republic of Congo is among the most vulnerable, insofar as it has limited room for adaptation, mainly due to poverty. Maintaining the services provided by natural ecosystems (forests, savannas, hydrological basins, etc.) is essential to ensure future development relays, limit the impacts of climate change and offer possibilities for adaptation to the most vulnerable groups, including are part of women and young people of all socio-cultural categories of urban and rural centers. - text: Finally, and from the generation of spaces for the exchange of experiences and good practices, specialists in the subject,. In terms of raising awareness, a discussion was held to integrate the gender perspective into the climate change agenda, which included views from government management and science. The discussion was aimed at the general public and had the participation of youth organizations that work to raise awareness and sensitize in the fight against climate change. - text: 'Over the past period, even the Party and the Government of Lao PDR were aware of the importance of and have paid attention to gender role. However, the status of women in Lao PDR in many fields is not equal to that of men, and women were still taken advantage of in many forms. Hence, in order to ensure that peoples of all gender and ages and all social strata are able to participate in the process and receive the benefits from the development in a comprehensive, inclusive and fair manner, the National Green Growth Strategy of the Lao PDR has identified gender role/protection and promotion of the advancement of women activities to be an important focus of the green growth and will particularly focus on: ' - text: Construction of pipelines and connection to existing ones to transmit water to demand centres. Reduce water loss during transmission by investing on telemetric monitoring systems. Enhance conjunctive groundwater-surface water use. Agriculture. Improve genetic characteristics of the livestock breed such as Musi breed. Improve livestock diet through supplementary feeding. A switch to crops with the following traits:. Drought resistant. Tolerant to high temperatures. Short maturity. Health. Public education and malaria campaigns. Malaria Strategy. Control of Diarrhoeal Diseases metrics: - accuracy pipeline_tag: text-classification library_name: setfit inference: false base_model: sentence-transformers/paraphrase-mpnet-base-v2 --- # SetFit with sentence-transformers/paraphrase-mpnet-base-v2 This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/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](https://www.sbert.net) 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](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2) - **Classification head:** a OneVsRestClassifier instance - **Maximum Sequence Length:** 256 tokens - **Number of Classes:** 17 classes ### Model Sources - **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit) - **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055) - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit) ## Uses ### Direct Use for Inference First install the SetFit library: ```bash pip install setfit ``` Then you can load this model and run inference. ```python from setfit import SetFitModel # Download from the 🤗 Hub model = SetFitModel.from_pretrained("peter2000/setfit-vulnerability-groups") # Run inference preds = model("The infrastructure requirement for collection based on the targets and projections made is presented in Table 12.22. A total of about 149,000 km length and 8,660km length of sewers are required for urban and rural communities, respectively by 2047. In addition, a little over 4 million facilities in urban areas and about 853,000 facilities in rural areas will be required to meet on-site sanitation needs by 2033 nationwide.") ``` ## Training Details ### Training Set Metrics | Training set | Min | Median | Max | |:-------------|:----|:--------|:----| | Word count | 15 | 71.2316 | 164 | ### Training Hyperparameters - batch_size: (16, 16) - num_epochs: (1, 1) - max_steps: -1 - sampling_strategy: oversampling - num_iterations: 20 - 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 - l2_weight: 0.01 - seed: 42 - eval_max_steps: -1 - load_best_model_at_end: False ### Training Results | Epoch | Step | Training Loss | Validation Loss | |:------:|:----:|:-------------:|:---------------:| | 0.0011 | 1 | 0.3006 | - | | 0.0526 | 50 | 0.2232 | - | | 0.1053 | 100 | 0.1670 | - | | 0.1579 | 150 | 0.1202 | - | | 0.2105 | 200 | 0.0935 | - | | 0.2632 | 250 | 0.0862 | - | | 0.3158 | 300 | 0.0626 | - | | 0.3684 | 350 | 0.0664 | - | | 0.4211 | 400 | 0.0555 | - | | 0.4737 | 450 | 0.0528 | - | | 0.5263 | 500 | 0.0543 | - | | 0.5789 | 550 | 0.0501 | - | | 0.6316 | 600 | 0.0535 | - | | 0.6842 | 650 | 0.0465 | - | | 0.7368 | 700 | 0.0468 | - | | 0.7895 | 750 | 0.0470 | - | | 0.8421 | 800 | 0.0421 | - | | 0.8947 | 850 | 0.0379 | - | | 0.9474 | 900 | 0.0475 | - | | 1.0 | 950 | 0.0449 | - | ### Framework Versions - Python: 3.12.12 - SetFit: 1.2.0 - Sentence Transformers: 6.1.0 - Transformers: 5.17.0 - PyTorch: 2.14.0+cu130 - Datasets: 5.0.1 - Tokenizers: 0.23.2 ## Citation ### BibTeX ```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} } ```