Text Classification
sentence-transformers
Safetensors
Transformers
multilingual
gemma
text-generation
Instructions to use BAAI/bge-reranker-v2-gemma with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use BAAI/bge-reranker-v2-gemma with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("BAAI/bge-reranker-v2-gemma") 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-reranker-v2-gemma with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="BAAI/bge-reranker-v2-gemma")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("BAAI/bge-reranker-v2-gemma") model = AutoModelForCausalLM.from_pretrained("BAAI/bge-reranker-v2-gemma") - Notebooks
- Google Colab
- Kaggle
about bge-reranker-gemma
#5
by gaven0319 - opened
why "score = bge_gemma_reranker.compute_score(["query", "passage"])" will output two score "[-5.078125, -2.0859375]", which socre should me use?
I run the following code, but get only one score, can you provide the complete code for more information
from FlagEmbedding import FlagLLMReranker
reranker = FlagLLMReranker('BAAI/bge-reranker-v2-gemma', use_fp16=True) # Setting use_fp16 to True speeds up computation with a slight performance degradation
score = reranker.compute_score(['query', 'passage'])
print(score) # [1.974609375]
Thx,I know what I did wrong, okay, I use FlagReranker. By the way, at what score can it be considered as a 'yes'?
Yes, the score is the logits of 'Yes'.
I have the same issue, can you please tell what the problem was? im using FlagEmbeddingReranker from Llama index and im getting two scores for each node :/