--- library_name: setfit tags: - setfit - absa - sentence-transformers - text-classification - generated_from_setfit_trainer base_model: firqaaa/indo-sentence-bert-large metrics: - accuracy widget: - text: gamenya seru bagus paket:gamenya seru bagus paket worth it gak lag mudah mainnya tugas hadiah bagus modenya sayangnya game kadang ngebug gapapa kasih - text: tolong perbaiki analog nya pengaturan posisi:tolong perbaiki analog nya pengaturan posisi berpindah pindah - text: visualisasi bagus segi graphic:visualisasi bagus segi graphic bagus ya game cocok sih mantra nya banyakin contoh mantra penghilang - text: jaringan udah bagus game jaringan nya bagus:game nya udah bagus jaringan game nya bermasalah jaringan udah bagus game jaringan nya bagus mohon nambahin karakter - text: kali game stuk loading server pakai jaringan:game bagus cma kendala kali game stuk loading server pakai jaringan wifi masuk jaringan jaringan bermasalah main game online lancar game susah akses tolong diperbaiki supercell detik bermain coc lancar masuk kendala pipeline_tag: text-classification inference: false model-index: - name: SetFit Polarity Model with firqaaa/indo-sentence-bert-large results: - task: type: text-classification name: Text Classification dataset: name: Unknown type: unknown split: test metrics: - type: accuracy value: 0.8677536231884058 name: Accuracy --- # SetFit Polarity Model with firqaaa/indo-sentence-bert-large This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Aspect Based Sentiment Analysis (ABSA). This SetFit model uses [firqaaa/indo-sentence-bert-large](https://huggingface.co/firqaaa/indo-sentence-bert-large) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification. In particular, this model is in charge of classifying aspect polarities. 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. This model was trained within the context of a larger system for ABSA, which looks like so: 1. Use a spaCy model to select possible aspect span candidates. 2. Use a SetFit model to filter these possible aspect span candidates. 3. **Use this SetFit model to classify the filtered aspect span candidates.** ## Model Details ### Model Description - **Model Type:** SetFit - **Sentence Transformer body:** [firqaaa/indo-sentence-bert-large](https://huggingface.co/firqaaa/indo-sentence-bert-large) - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance - **spaCy Model:** id_core_news_trf - **SetFitABSA Aspect Model:** [Funnyworld1412/ABSA_indo-sentence-bert-large_MiniLM-L6-aspect](https://huggingface.co/Funnyworld1412/ABSA_indo-sentence-bert-large_MiniLM-L6-aspect) - **SetFitABSA Polarity Model:** [Funnyworld1412/ABSA_indo-sentence-bert-large_MiniLM-L6-polarity](https://huggingface.co/Funnyworld1412/ABSA_indo-sentence-bert-large_MiniLM-L6-polarity) - **Maximum Sequence Length:** 512 tokens - **Number of Classes:** 2 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) ### Model Labels | Label | Examples | |:--------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | negatif | | | positif | | ## Evaluation ### Metrics | Label | Accuracy | |:--------|:---------| | **all** | 0.8678 | ## 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 AbsaModel # Download from the 🤗 Hub model = AbsaModel.from_pretrained( "Funnyworld1412/ABSA_indo-sentence-bert-large_MiniLM-L6-aspect", "Funnyworld1412/ABSA_indo-sentence-bert-large_MiniLM-L6-polarity", ) # Run inference preds = model("The food was great, but the venue is just way too busy.") ``` ## Training Details ### Training Set Metrics | Training set | Min | Median | Max | |:-------------|:----|:--------|:----| | Word count | 3 | 28.3626 | 83 | | Label | Training Sample Count | |:--------|:----------------------| | negatif | 738 | | positif | 528 | ### Training Hyperparameters - batch_size: (4, 4) - num_epochs: (1, 1) - max_steps: -1 - sampling_strategy: oversampling - num_iterations: 5 - 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 - seed: 42 - eval_max_steps: -1 - load_best_model_at_end: False ### Training Results | Epoch | Step | Training Loss | Validation Loss | |:------:|:----:|:-------------:|:---------------:| | 0.0003 | 1 | 0.2752 | - | | 0.0158 | 50 | 0.2227 | - | | 0.0316 | 100 | 0.2989 | - | | 0.0474 | 150 | 0.2169 | - | | 0.0632 | 200 | 0.2081 | - | | 0.0790 | 250 | 0.0653 | - | | 0.0948 | 300 | 0.3878 | - | | 0.1106 | 350 | 0.0675 | - | | 0.1264 | 400 | 0.0096 | - | | 0.1422 | 450 | 0.2128 | - | | 0.1580 | 500 | 0.028 | - | | 0.1738 | 550 | 0.177 | - | | 0.1896 | 600 | 0.006 | - | | 0.2054 | 650 | 0.0159 | - | | 0.2212 | 700 | 0.2219 | - | | 0.2370 | 750 | 0.0085 | - | | 0.2528 | 800 | 0.0202 | - | | 0.2686 | 850 | 0.0174 | - | | 0.2844 | 900 | 0.1191 | - | | 0.3002 | 950 | 0.0645 | - | | 0.3160 | 1000 | 0.004 | - | | 0.3318 | 1050 | 0.0132 | - | | 0.3476 | 1100 | 0.0036 | - | | 0.3633 | 1150 | 0.0056 | - | | 0.3791 | 1200 | 0.0074 | - | | 0.3949 | 1250 | 0.0012 | - | | 0.4107 | 1300 | 0.0024 | - | | 0.4265 | 1350 | 0.0028 | - | | 0.4423 | 1400 | 0.0011 | - | | 0.4581 | 1450 | 0.1069 | - | | 0.4739 | 1500 | 0.0013 | - | | 0.4897 | 1550 | 0.0006 | - | | 0.5055 | 1600 | 0.0012 | - | | 0.5213 | 1650 | 0.1649 | - | | 0.5371 | 1700 | 0.0019 | - | | 0.5529 | 1750 | 0.0016 | - | | 0.5687 | 1800 | 0.2247 | - | | 0.5845 | 1850 | 0.0019 | - | | 0.6003 | 1900 | 0.0131 | - | | 0.6161 | 1950 | 0.0008 | - | | 0.6319 | 2000 | 0.0006 | - | | 0.6477 | 2050 | 0.0008 | - | | 0.6635 | 2100 | 0.0011 | - | | 0.6793 | 2150 | 0.1962 | - | | 0.6951 | 2200 | 0.0007 | - | | 0.7109 | 2250 | 0.0007 | - | | 0.7267 | 2300 | 0.0009 | - | | 0.7425 | 2350 | 0.0007 | - | | 0.7583 | 2400 | 0.0006 | - | | 0.7741 | 2450 | 0.0015 | - | | 0.7899 | 2500 | 0.0005 | - | | 0.8057 | 2550 | 0.0007 | - | | 0.8215 | 2600 | 0.0016 | - | | 0.8373 | 2650 | 0.0008 | - | | 0.8531 | 2700 | 0.0013 | - | | 0.8689 | 2750 | 0.0007 | - | | 0.8847 | 2800 | 0.0008 | - | | 0.9005 | 2850 | 0.0008 | - | | 0.9163 | 2900 | 0.0005 | - | | 0.9321 | 2950 | 0.0006 | - | | 0.9479 | 3000 | 0.0006 | - | | 0.9637 | 3050 | 0.0006 | - | | 0.9795 | 3100 | 0.0004 | - | | 0.9953 | 3150 | 0.0005 | - | | 1.0 | 3165 | - | 0.2012 | ### Framework Versions - Python: 3.10.13 - SetFit: 1.0.3 - Sentence Transformers: 3.0.1 - spaCy: 3.7.5 - Transformers: 4.36.2 - PyTorch: 2.1.2 - Datasets: 2.19.2 - Tokenizers: 0.15.2 ## 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} } ```