faodl/model_g20_multilabel
08
1---2tags:3- setfit4- sentence-transformers5- text-classification6- generated_from_setfit_trainer7widget:8- text: "6) Implement advocacy strategies with heads of government ministries, departments\9 \ \n\nand institutions, national, district and local leaders on solutions to major\10 \ nutrition \nproblems."11- text: 'The Government plans to continue its interventions aimed at increasing access12 to drinking water by:13 14 - in rural areas, constructing an additional 2 500 water points (mainly boreholes)15 and rehabilitating an extra 2 000 existing water points.16 17 - in urban and pre-urban areas, rehabilitating and constructing water supply infrastructure18 in the various urban towns.19 20 The Government will also, in terms of sanitation, continue to promote community-based21 approaches and construct facilities.22 23 24 Objective: ensure adequate access to sanitation facilities and increase access25 to clean and safe drinking water from 64% (2014) to 67% of the population in urban26 areas and from 83% (2014) to 85% in urban and pre-urban areas.27 28 29 '30- text: "Specific objective\ni) to improve safe water supply services to the people\31 \ in the rural communities\nii) to improve the water supply service levels in\32 \ rural area to enable rural the population in the \nproject areas to increase\33 \ their economic income through incorporating back yard or mini \nirrigation system."34- text: "Social security contributions \n\nLabor \nMarkets \n\nActivation measures\35 \ \n\nā¢ During the period of state of emergency, all training activities \nrecognized\36 \ by the Ministry of Labor and Social Protection can be \ndelivered online."37- text: "Training infrastructure will be adapted to accommodate \nnew\tprogrammes."38metrics:39- accuracy40pipeline_tag: text-classification41library_name: setfit42inference: false43base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v244---45 46# SetFit with sentence-transformers/paraphrase-multilingual-MiniLM-L12-v247 48This 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.49 50The model has been trained using an efficient few-shot learning technique that involves:51 521. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.532. Training a classification head with features from the fine-tuned Sentence Transformer.54 55## Model Details56 57### Model Description58- **Model Type:** SetFit59- **Sentence Transformer body:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2)60- **Classification head:** a OneVsRestClassifier instance61- **Maximum Sequence Length:** 128 tokens62<!-- - **Number of Classes:** Unknown -->63<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->64<!-- - **Language:** Unknown -->65<!-- - **License:** Unknown -->66 67### Model Sources68 69- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)70- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)71- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)72 73## Uses74 75### Direct Use for Inference76 77First install the SetFit library:78 79```bash80pip install setfit81```82 83Then you can load this model and run inference.84 85```python86from setfit import SetFitModel87 88# Download from the š¤ Hub89model = SetFitModel.from_pretrained("faodl/model_g20_multilabel")90# Run inference91preds = model("Training infrastructure will be adapted to accommodate 92new programmes.")93```94 95<!--96### Downstream Use97 98*List how someone could finetune this model on their own dataset.*99-->100 101<!--102### Out-of-Scope Use103 104*List how the model may foreseeably be misused and address what users ought not to do with the model.*105-->106 107<!--108## Bias, Risks and Limitations109 110*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*111-->112 113<!--114### Recommendations115 116*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*117-->118 119## Training Details120 121### Training Set Metrics122| Training set | Min | Median | Max |123|:-------------|:----|:--------|:-----|124| Word count | 1 | 48.9866 | 1181 |125 126### Training Hyperparameters127- batch_size: (16, 16)128- num_epochs: (1, 1)129- max_steps: -1130- sampling_strategy: oversampling131- num_iterations: 50132- body_learning_rate: (2e-05, 2e-05)133- head_learning_rate: 2e-05134- loss: CosineSimilarityLoss135- distance_metric: cosine_distance136- margin: 0.25137- end_to_end: False138- use_amp: False139- warmup_proportion: 0.1140- l2_weight: 0.01141- seed: 42142- eval_max_steps: -1143- load_best_model_at_end: False144 145### Training Results146| Epoch | Step | Training Loss | Validation Loss |147|:------:|:----:|:-------------:|:---------------:|148| 0.0002 | 1 | 0.2348 | - |149| 0.0119 | 50 | 0.1747 | - |150| 0.0237 | 100 | 0.153 | - |151| 0.0356 | 150 | 0.1314 | - |152| 0.0475 | 200 | 0.1263 | - |153| 0.0593 | 250 | 0.1168 | - |154| 0.0712 | 300 | 0.116 | - |155| 0.0831 | 350 | 0.098 | - |156| 0.0949 | 400 | 0.1085 | - |157| 0.1068 | 450 | 0.0975 | - |158| 0.1187 | 500 | 0.094 | - |159| 0.1305 | 550 | 0.082 | - |160| 0.1424 | 600 | 0.0856 | - |161| 0.1543 | 650 | 0.0838 | - |162| 0.1662 | 700 | 0.0762 | - |163| 0.1780 | 750 | 0.0722 | - |164| 0.1899 | 800 | 0.0722 | - |165| 0.2018 | 850 | 0.0634 | - |166| 0.2136 | 900 | 0.0584 | - |167| 0.2255 | 950 | 0.0664 | - |168| 0.2374 | 1000 | 0.0688 | - |169| 0.2492 | 1050 | 0.0629 | - |170| 0.2611 | 1100 | 0.0579 | - |171| 0.2730 | 1150 | 0.0652 | - |172| 0.2848 | 1200 | 0.0573 | - |173| 0.2967 | 1250 | 0.0584 | - |174| 0.3086 | 1300 | 0.0558 | - |175| 0.3204 | 1350 | 0.0586 | - |176| 0.3323 | 1400 | 0.0574 | - |177| 0.3442 | 1450 | 0.0444 | - |178| 0.3560 | 1500 | 0.0462 | - |179| 0.3679 | 1550 | 0.0488 | - |180| 0.3798 | 1600 | 0.0505 | - |181| 0.3916 | 1650 | 0.0529 | - |182| 0.4035 | 1700 | 0.0487 | - |183| 0.4154 | 1750 | 0.0459 | - |184| 0.4272 | 1800 | 0.0531 | - |185| 0.4391 | 1850 | 0.0448 | - |186| 0.4510 | 1900 | 0.0382 | - |187| 0.4629 | 1950 | 0.0457 | - |188| 0.4747 | 2000 | 0.0493 | - |189| 0.4866 | 2050 | 0.0488 | - |190| 0.4985 | 2100 | 0.049 | - |191| 0.5103 | 2150 | 0.0495 | - |192| 0.5222 | 2200 | 0.0402 | - |193| 0.5341 | 2250 | 0.0493 | - |194| 0.5459 | 2300 | 0.0496 | - |195| 0.5578 | 2350 | 0.0438 | - |196| 0.5697 | 2400 | 0.0361 | - |197| 0.5815 | 2450 | 0.0428 | - |198| 0.5934 | 2500 | 0.0419 | - |199| 0.6053 | 2550 | 0.0416 | - |200| 0.6171 | 2600 | 0.0338 | - |201| 0.6290 | 2650 | 0.0397 | - |202| 0.6409 | 2700 | 0.0385 | - |203| 0.6527 | 2750 | 0.0285 | - |204| 0.6646 | 2800 | 0.0461 | - |205| 0.6765 | 2850 | 0.0341 | - |206| 0.6883 | 2900 | 0.0379 | - |207| 0.7002 | 2950 | 0.0435 | - |208| 0.7121 | 3000 | 0.0341 | - |209| 0.7239 | 3050 | 0.0395 | - |210| 0.7358 | 3100 | 0.0424 | - |211| 0.7477 | 3150 | 0.0415 | - |212| 0.7596 | 3200 | 0.0422 | - |213| 0.7714 | 3250 | 0.0402 | - |214| 0.7833 | 3300 | 0.0309 | - |215| 0.7952 | 3350 | 0.0379 | - |216| 0.8070 | 3400 | 0.039 | - |217| 0.8189 | 3450 | 0.0427 | - |218| 0.8308 | 3500 | 0.0331 | - |219| 0.8426 | 3550 | 0.0457 | - |220| 0.8545 | 3600 | 0.0306 | - |221| 0.8664 | 3650 | 0.034 | - |222| 0.8782 | 3700 | 0.0354 | - |223| 0.8901 | 3750 | 0.0393 | - |224| 0.9020 | 3800 | 0.036 | - |225| 0.9138 | 3850 | 0.0339 | - |226| 0.9257 | 3900 | 0.0332 | - |227| 0.9376 | 3950 | 0.0274 | - |228| 0.9494 | 4000 | 0.0372 | - |229| 0.9613 | 4050 | 0.0319 | - |230| 0.9732 | 4100 | 0.0339 | - |231| 0.9850 | 4150 | 0.0349 | - |232| 0.9969 | 4200 | 0.0383 | - |233 234### Framework Versions235- Python: 3.11.13236- SetFit: 1.1.2237- Sentence Transformers: 4.1.0238- Transformers: 4.52.4239- PyTorch: 2.6.0+cu124240- Datasets: 3.6.0241- Tokenizers: 0.21.1242 243## Citation244 245### BibTeX246```bibtex247@article{https://doi.org/10.48550/arxiv.2209.11055,248 doi = {10.48550/ARXIV.2209.11055},249 url = {https://arxiv.org/abs/2209.11055},250 author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},251 keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},252 title = {Efficient Few-Shot Learning Without Prompts},253 publisher = {arXiv},254 year = {2022},255 copyright = {Creative Commons Attribution 4.0 International}256}257```258 259<!--260## Glossary261 262*Clearly define terms in order to be accessible across audiences.*263-->264 265<!--266## Model Card Authors267 268*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*269-->270 271<!--272## Model Card Contact273 274*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*275-->