--- tags: - setfit - sentence-transformers - text-classification - generated_from_setfit_trainer widget: - text: To monitor market dynamics and inform policy responses, the government will track the retail value of ultra-processed foods and analyze shifts in consumption in relation to labeling and advertising reforms. Data from these analyses will feed annual dashboards that link labeling density, promotional intensity, and dietary outcomes to guide targeted interventions and budget planning. - text: the national agricultural plan is a national sectoral plan of grenada of 2015-2030. its main goal is to stimulate economic growth in the agriculture sector through the development of a well-coordinated planning and implementation framework that is interactive and effective, and involve the full participation of the stakeholders, and which promotes food security, income generation and poverty alleviation. in the area of food security, the document aims to reduce dependence on food imports and imported staples in particular and increase availability of local fresh and fresh processed products; increase economic access to food by vulnerable persons and their capacity to address their food and nutrition needs; and to improve the health status and wellbeing of the grenadians through the consumption of nutritious and safe foods. the plan also seeks to make agriculture, forestry and fisheries more productive and sustainable. specifically, it envisions to build climate resilience to avoid, prevent, or minimize climate change impacts on agriculture (including forestry and fisheries), the environment and biodiversity; improve preparedness for climate change impacts and extreme events; enhance the country’s response capacity in case of extremes; facilitate recovery from impacts and extremes; and reduce the impact of land based agriculture on climate change and the environment; and preserve and optimize resources (land, sea, genetic). moreover, the document aims to reduce rural poverty. in particular, it provides for making additional investments in economic infrastructure for increased contribution of the agricultural sector to economic growth, poverty alleviation and environmental sustainability. further, the plan targets to increase exports of traditional crops, fish, fruits, vegetables, root crops, minor spices, and value added products to international and regional markets; increase production of targeted fruits, vegetables, root crops, herbs and minor spices for targeted domestic markets; make additional investments in institutional and human resource capacity development in the agricultural sector to improve governance and efficiency; achieve greater collaboration in regional and international trade for agricultural products; create framework for donor and development partner coordination in providing support for the agriculture sector; leverage opportunities in the tourism sector to strengthen the linkage between agriculture and tourism; and invest in upgrading agricultural research and development capacity. institutional responsibility for the implementation of the plan is with the ministry of agriculture, lands, forestry, fisheries and the environment. the minister will be obligated to report to the cabinet and parliament on progress in the implementation of the plan. it is expected that the plan will be incorporated into the national sustainable development plan 2030 (nsdp2030). the ministry through the permanent secretary will be expected to report to the monitoring committee of the nsdp2030 on a monthly basis on progress in implementation. the reports to the cabinet will be submitted biannually. - text: 'the seven key objectives are: 1. improve coordination in the sector to successfully implement the fruit and vegetable strategy 2. improve market intelligence, promotion and dissemination across the whole value chain 3. build a supply sub sector that can guarantee consistent quality and supply of fresh fruit and vegetables 4. build a sector that is well trained and supported by a comprehensive and properly executed capability plan 5. improve financial situation of sector farmers and enterprises 6. promote integrated management of resources to ensure sustainability of the fruit and vegetable sector 7. strengthen samoa association for manufacturers and exporters (same) to provide services that will increase returns and overall value addition for sector' - text: Trade facilitation should be aligned with nutrition security and rural development by prioritizing critical food and input imports, harmonizing rules of origin with neighboring economies, and strengthening transit corridors to support small producers. Progress indicators include the ratio of food imports to merchandise imports and the share of agricultural raw materials imports, alongside the incidence of firms naming customs and trade regulations as top obstacles (6.6.3.3). - text: 1. general objectives striving to be a developing country with modern industry and high middle income by 2030; have a modern, competitive, effective and effective management institution; the economy develops dynamically, quickly and sustainably, independently and autonomously on the basis of science, technology and innovation in association with improving efficiency in external activities and international integration; arousing the aspiration to develop the country, promoting the creativity, will and strength of the whole nation, building a prosperous, democratic, fair, civilized, orderly, disciplined and safe society, ensuring a peaceful and happy life of the people; constantly improve all aspects of people's lives; firmly protect the fatherland, a peaceful and stable environment for national development; improve vietnam's position and prestige in the international arena. striving to become a developed and high-income country by 2045. 2. principal indicators a) regarding the economy - the average growth rate of gross domestic product (gdp) is about 7%/year; gdp per capita at current prices by 2030 will reach about 7,500 usd3. - the proportion of the processing and manufacturing industry will reach about 30% of gdp, and the digital economy will reach about 30% of gdp. - the urbanization rate will reach over 50%. - the average total social investment will reach 33-35% of gdp; public debt does not exceed 60% of gdp. - the contribution of total factor productivity (tfp) to growth reached 50%. - the average growth rate of social labor productivity will reach over 6.5%/year. - reduce energy consumption per unit of gdp at 1-1.5%/year. b) regarding social - the human development index (hdi) remained above 0.74. - the average life expectancy is 75 years, of which the healthy life span is at least 68 years. - the percentage of trained workers with degrees and certificates reaches 35-40%. - the proportion of agricultural labor in the total social labor force will decrease to less than 20%. c) regarding the environment - the forest cover rate is stable at 42%. - the rate of treatment and reuse of wastewater into the river basin environment will reach over 70%. - reduce greenhouse gas emissions by 9%5. - 100% of production and business establishments meet environmental standards. - to increase the area of marine and coastal protected areas to 3-5% of the natural area of national waters. metrics: - accuracy pipeline_tag: text-classification library_name: setfit inference: false base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 model-index: - name: SetFit with sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 results: - task: type: text-classification name: Text Classification dataset: name: Unknown type: unknown split: test metrics: - type: accuracy value: 0.2326797385620915 name: Accuracy --- # SetFit with sentence-transformers/paraphrase-multilingual-MiniLM-L12-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-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-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-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) - **Classification head:** a OneVsRestClassifier instance - **Maximum Sequence Length:** 128 tokens ### 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) ## Evaluation ### Metrics | Label | Accuracy | |:--------|:---------| | **all** | 0.2327 | ## 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("faodl/model_cca_multilabel_MiniLM-L12-v01") # Run inference preds = model("To monitor market dynamics and inform policy responses, the government will track the retail value of ultra-processed foods and analyze shifts in consumption in relation to labeling and advertising reforms. Data from these analyses will feed annual dashboards that link labeling density, promotional intensity, and dietary outcomes to guide targeted interventions and budget planning.") ``` ## Training Details ### Training Set Metrics | Training set | Min | Median | Max | |:-------------|:----|:---------|:----| | Word count | 1 | 123.6200 | 951 | ### Training Hyperparameters - batch_size: (32, 32) - num_epochs: (2, 2) - 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 - 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 | 0.1892 | - | | 0.0566 | 50 | 0.192 | - | | 0.1131 | 100 | 0.1681 | - | | 0.1697 | 150 | 0.1518 | - | | 0.2262 | 200 | 0.1361 | - | | 0.2828 | 250 | 0.1389 | - | | 0.3394 | 300 | 0.1321 | - | | 0.3959 | 350 | 0.1297 | - | | 0.4525 | 400 | 0.1236 | - | | 0.5090 | 450 | 0.1116 | - | | 0.5656 | 500 | 0.1194 | - | | 0.6222 | 550 | 0.1105 | - | | 0.6787 | 600 | 0.1047 | - | | 0.7353 | 650 | 0.1124 | - | | 0.7919 | 700 | 0.1069 | - | | 0.8484 | 750 | 0.108 | - | | 0.9050 | 800 | 0.1072 | - | | 0.9615 | 850 | 0.1011 | - | | 1.0181 | 900 | 0.098 | - | | 1.0747 | 950 | 0.0893 | - | | 1.1312 | 1000 | 0.0979 | - | | 1.1878 | 1050 | 0.0967 | - | | 1.2443 | 1100 | 0.0887 | - | | 1.3009 | 1150 | 0.0908 | - | | 1.3575 | 1200 | 0.0906 | - | | 1.4140 | 1250 | 0.0869 | - | | 1.4706 | 1300 | 0.0873 | - | | 1.5271 | 1350 | 0.0943 | - | | 1.5837 | 1400 | 0.0886 | - | | 1.6403 | 1450 | 0.0911 | - | | 1.6968 | 1500 | 0.0832 | - | | 1.7534 | 1550 | 0.0859 | - | | 1.8100 | 1600 | 0.0862 | - | | 1.8665 | 1650 | 0.09 | - | | 1.9231 | 1700 | 0.0836 | - | | 1.9796 | 1750 | 0.0884 | - | | 0.0006 | 1 | 0.0898 | - | | 0.0283 | 50 | 0.09 | - | | 0.0566 | 100 | 0.091 | - | | 0.0849 | 150 | 0.0905 | - | | 0.1132 | 200 | 0.085 | - | | 0.1415 | 250 | 0.0862 | - | | 0.1698 | 300 | 0.0915 | - | | 0.1981 | 350 | 0.0865 | - | | 0.2264 | 400 | 0.0873 | - | | 0.2547 | 450 | 0.0897 | - | | 0.2830 | 500 | 0.0906 | - | | 0.3113 | 550 | 0.096 | - | | 0.3396 | 600 | 0.0886 | - | | 0.3679 | 650 | 0.0831 | - | | 0.3962 | 700 | 0.0852 | - | | 0.4244 | 750 | 0.0858 | - | | 0.4527 | 800 | 0.0831 | - | | 0.4810 | 850 | 0.0858 | - | | 0.5093 | 900 | 0.0898 | - | | 0.5376 | 950 | 0.0866 | - | | 0.5659 | 1000 | 0.0836 | - | | 0.5942 | 1050 | 0.0809 | - | | 0.6225 | 1100 | 0.0838 | - | | 0.6508 | 1150 | 0.0845 | - | | 0.6791 | 1200 | 0.0803 | - | | 0.7074 | 1250 | 0.0831 | - | | 0.7357 | 1300 | 0.0799 | - | | 0.7640 | 1350 | 0.0853 | - | | 0.7923 | 1400 | 0.0786 | - | | 0.8206 | 1450 | 0.0763 | - | | 0.8489 | 1500 | 0.0795 | - | | 0.8772 | 1550 | 0.08 | - | | 0.9055 | 1600 | 0.0786 | - | | 0.9338 | 1650 | 0.0759 | - | | 0.9621 | 1700 | 0.0817 | - | | 0.9904 | 1750 | 0.0712 | - | | 1.0187 | 1800 | 0.0703 | - | | 1.0470 | 1850 | 0.0702 | - | | 1.0753 | 1900 | 0.0704 | - | | 1.1036 | 1950 | 0.0759 | - | | 1.1319 | 2000 | 0.0716 | - | | 1.1602 | 2050 | 0.0714 | - | | 1.1885 | 2100 | 0.0698 | - | | 1.2168 | 2150 | 0.0734 | - | | 1.2450 | 2200 | 0.0717 | - | | 1.2733 | 2250 | 0.0671 | - | | 1.3016 | 2300 | 0.0681 | - | | 1.3299 | 2350 | 0.072 | - | | 1.3582 | 2400 | 0.0685 | - | | 1.3865 | 2450 | 0.0702 | - | | 1.4148 | 2500 | 0.0673 | - | | 1.4431 | 2550 | 0.0698 | - | | 1.4714 | 2600 | 0.0667 | - | | 1.4997 | 2650 | 0.0658 | - | | 1.5280 | 2700 | 0.0759 | - | | 1.5563 | 2750 | 0.067 | - | | 1.5846 | 2800 | 0.0777 | - | | 1.6129 | 2850 | 0.0699 | - | | 1.6412 | 2900 | 0.0773 | - | | 1.6695 | 2950 | 0.0704 | - | | 1.6978 | 3000 | 0.0731 | - | | 1.7261 | 3050 | 0.0682 | - | | 1.7544 | 3100 | 0.0684 | - | | 1.7827 | 3150 | 0.0628 | - | | 1.8110 | 3200 | 0.0689 | - | | 1.8393 | 3250 | 0.068 | - | | 1.8676 | 3300 | 0.0652 | - | | 1.8959 | 3350 | 0.0714 | - | | 1.9242 | 3400 | 0.0714 | - | | 1.9525 | 3450 | 0.0701 | - | | 1.9808 | 3500 | 0.0644 | - | ### Framework Versions - Python: 3.12.12 - SetFit: 1.1.3 - Sentence Transformers: 5.1.1 - Transformers: 4.57.1 - PyTorch: 2.8.0+cu126 - Datasets: 4.0.0 - Tokenizers: 0.22.1 ## 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} } ```