--- library_name: setfit tags: - setfit - sentence-transformers - text-classification - generated_from_setfit_trainer base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 metrics: - accuracy widget: - text: Mitarbeit am wissenschaftlichen Arbeitsplatz (LV0125) - text: '20-00-0366-iv Serious Games: Einführung in die Thematik „Serious Games“: wissenschaftlich-technische Grundlagen, Anwendungsgebiete und Trends. Die Einzelthemen umfassen unter anderem: • Einführung in Serious Games • Game Development, Game Design • Game Technology, Tools und Engines • Personalisierung und Adaption • Interactive Digital Storytelling • Authoring und Content Generation • Multiplayer Games • Game Interfaces und Sensor Technology • Effects, Affects und User Experience • Mobile Games • Serious Games Anwendungsbereiche und Best-Practice Beispiele Die Übungen enthalten Theorie- und Praxisanteile. Dabei wird die Verwendung einer Game Engine gelehrt.' - text: 'Aerobotics Seminar: - Einführung in die Aufgabenstellung, die vorhandene Infrastruktur und den zu durchlaufenden Entwicklungsprozess - Entwurf und Implementierung von Algorithmen zur Flugregelung in Gruppenarbeit - Diskussion des Fortschritts in regelmäßigen Progress-Meetings - Flugdemonstration - Abschließende Präsentation und Dokumentation' - text: "Digital Transformations, Consumer Well-Being, and Sustainability: Physical\ \ consumption, and individual ownership of material products in particular, has\ \ traditionally been the default mode of consumption and its extent has long been\ \ considered a measure of personal and societal prosperity. However, our daily\ \ lives increasingly shift towards or are altered by digital environments, which\ \ nurture alternative forms of consumption such as sharing or access-based offers,\ \ cultivate the prevalence of virtual living worlds through fictional experiences,\ \ and alter our relations to material possessions by an increasing availability\ \ of digital solutions.\n\n\n\nThe course exposes you to state-of-the-art research\ \ on consumer research and digital transformations in fields such as virtual reality,\ \ sharing economy, and blockchain technologies. You will be guided through background\ \ information of consumer behavior and consumer psychology. By creating a collaborative\ \ learning environment, we will explore and critically discuss how digital transformations\ \ affect consumer well-being and sustainability. \n\n\n\nYour role is to be an\ \ active contributor in the class. This course consists of a lectures, discussion\ \ sessions and group presentations. Generally, analysis of readings will be used\ \ to guide our discussion. \n\n\n\nThe main objective of the course is to critically\ \ reflect upon current technological advancements that increasingly permeate everyday\ \ lives. Students will be engaged in exploring technological-social issues in\ \ marketing and be guided into a critical approach on technology-brands-consumers\ \ relationships behind digital transformations." - text: 'Grundlagen der IT-Sicherheit: Um einen Überblick der IT-Sicherheit zu vermitteln werden folgende Themen behandelt: Motivation für IT-Sicherheit Grundbegriffe der IT-Sicherheit Computer Malware Kryptographische Grundlagen Authentisierung Biometrie Zugriffskontrolle Netzwerkund Internetsicherheit Physikalische Sicherheit / Physikalische Angriffe Sicherheitsevaluierung und Zertifizierung Einführung in den Datenschutz' pipeline_tag: text-classification inference: false --- # 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) ## 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("Chernoffface/fs-setfit-multilable-model") # Run inference preds = model("Mitarbeit am wissenschaftlichen Arbeitsplatz (LV0125)") ``` ## Training Details ### Training Set Metrics | Training set | Min | Median | Max | |:-------------|:----|:---------|:----| | Word count | 1 | 120.1215 | 537 | ### Training Hyperparameters - batch_size: (16, 16) - 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 | 0.2964 | - | | 0.0552 | 50 | 0.2042 | - | | 0.1105 | 100 | 0.1643 | - | | 0.1657 | 150 | 0.1376 | - | | 0.2210 | 200 | 0.1232 | - | | 0.2762 | 250 | 0.1125 | - | | 0.3315 | 300 | 0.1079 | - | | 0.3867 | 350 | 0.0951 | - | | 0.4420 | 400 | 0.0847 | - | | 0.4972 | 450 | 0.0917 | - | | 0.5525 | 500 | 0.085 | - | | 0.6077 | 550 | 0.0758 | - | | 0.6630 | 600 | 0.0743 | - | | 0.7182 | 650 | 0.0671 | - | | 0.7735 | 700 | 0.0743 | - | | 0.8287 | 750 | 0.0571 | - | | 0.8840 | 800 | 0.0625 | - | | 0.9392 | 850 | 0.0607 | - | | 0.9945 | 900 | 0.0686 | - | | 1.0497 | 950 | 0.0541 | - | | 1.1050 | 1000 | 0.0553 | - | | 1.1602 | 1050 | 0.0565 | - | | 1.2155 | 1100 | 0.0558 | - | | 1.2707 | 1150 | 0.0578 | - | | 1.3260 | 1200 | 0.0525 | - | | 1.3812 | 1250 | 0.0541 | - | | 1.4365 | 1300 | 0.049 | - | | 1.4917 | 1350 | 0.0485 | - | | 1.5470 | 1400 | 0.0475 | - | | 1.6022 | 1450 | 0.0479 | - | | 1.6575 | 1500 | 0.0514 | - | | 1.7127 | 1550 | 0.0509 | - | | 1.7680 | 1600 | 0.0517 | - | | 1.8232 | 1650 | 0.0455 | - | | 1.8785 | 1700 | 0.0493 | - | | 1.9337 | 1750 | 0.0501 | - | | 1.9890 | 1800 | 0.0492 | - | ### Framework Versions - Python: 3.12.3 - SetFit: 1.1.0 - Sentence Transformers: 3.0.0 - Transformers: 4.43.1 - PyTorch: 2.3.1+cu121 - Datasets: 2.20.0 - Tokenizers: 0.19.1 ## 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} } ```