Feature Extraction
sentence-transformers
Safetensors
Transformers
gemma2
sentence-similarity
mteb
Eval Results (legacy)
Instructions to use BAAI/bge-multilingual-gemma2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use BAAI/bge-multilingual-gemma2 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("BAAI/bge-multilingual-gemma2") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Transformers
How to use BAAI/bge-multilingual-gemma2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="BAAI/bge-multilingual-gemma2")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("BAAI/bge-multilingual-gemma2") model = AutoModel.from_pretrained("BAAI/bge-multilingual-gemma2") - Inference
- Notebooks
- Google Colab
- Kaggle
在PyTorch 2.1.0的环境中跑样例结果差异过大
#6
by shizue - opened
推理方式:transformers
官方样例提供的结果为:[[55.92064666748047, 1.6549524068832397], [-0.2698777914047241, 49.95653533935547]]
本地推理环境:cuda 12.1,pytorch 2.1.0
推理结果:[[55.93535232543945, 1.679487943649292], [-0.20362739264965057, 49.97922897338867]]
根据我的测试在 pytorch 版本升到 2.3.1 后结果符合预期,但在 2.1.0 下计算出的 score 和提供的参考值最大差出来 1/4 了(-0.26和-0.20),这可能会是什么原因导致的呢?