Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Horizon-Labs
/
multilingual-zeroshot-base

Zero-Shot Classification
Transformers
ONNX
Safetensors
Transformers.js
modernbert
text-classification
zero-shot
nli
natural-language-inference
mmbert
Model card Files Files and versions
xet
Community

Instructions to use Horizon-Labs/multilingual-zeroshot-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

  • Libraries
  • Transformers

    How to use Horizon-Labs/multilingual-zeroshot-base with Transformers:

    # Use a pipeline as a high-level helper
    from transformers import pipeline
    
    pipe = pipeline("zero-shot-classification", model="Horizon-Labs/multilingual-zeroshot-base")
    # pip install -U transformers accelerate
    # Load model directly
    from transformers import AutoTokenizer, AutoModelForSequenceClassification
    
    tokenizer = AutoTokenizer.from_pretrained("Horizon-Labs/multilingual-zeroshot-base")
    model = AutoModelForSequenceClassification.from_pretrained("Horizon-Labs/multilingual-zeroshot-base", device_map="auto")
  • Transformers.js

    How to use Horizon-Labs/multilingual-zeroshot-base with Transformers.js:

    // npm i @huggingface/transformers
    import { pipeline } from '@huggingface/transformers';
    
    // Allocate pipeline
    const pipe = await pipeline('zero-shot-classification', 'Horizon-Labs/multilingual-zeroshot-base');
  • Notebooks
  • Google Colab
  • Kaggle
multilingual-zeroshot-base / code /zeroshot
49 kB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 6 commits
Horizon-Labs's picture
Horizon-Labs
Card: two more unseen-label benchmarks (MTOP 6 languages, CLINC150)
8ec98ab verified about 1 month ago
  • build_evals.py
    4.6 kB
    Release multilingual zero-shot classifier v1.0 (base) about 1 month ago
  • build_evals2.py
    2.34 kB
    Card: two more unseen-label benchmarks (MTOP 6 languages, CLINC150) about 1 month ago
  • build_zs_data.py
    6.63 kB
    v1.3: native-language labels + distillation (native +0.021 MASSIVE / +0.023 SIB-200; B77 .568, SIB .826) about 1 month ago
  • evaluate_zs.py
    4.7 kB
    Release multilingual zero-shot classifier v1.0 (base) about 1 month ago
  • gen_zeroshot.py
    6.7 kB
    Release multilingual zero-shot classifier v1.0 (base) about 1 month ago
  • gen_zeroshot_v2.py
    12.9 kB
    Add v1.1 data generator code about 1 month ago
  • score_soft.py
    1.68 kB
    Add distillation scoring code about 1 month ago
  • translate_labels.py
    2.82 kB
    v1.3: native-language labels + distillation (native +0.021 MASSIVE / +0.023 SIB-200; B77 .568, SIB .826) about 1 month ago
  • translate_labelsets.py
    2.89 kB
    v1.3: native-language labels + distillation (native +0.021 MASSIVE / +0.023 SIB-200; B77 .568, SIB .826) about 1 month ago
  • translate_nli.py
    3.75 kB
    Release multilingual zero-shot classifier v1.0 (base) about 1 month ago