RikoteMaster/MNLP_M3_document_encoder

1---2license: apache-2.03base_model: ibm-granite/granite-embedding-107m-multilingual4tags:5- sentence-transformers6- feature-extraction7- sentence-similarity8- transformers9- granite10- embeddings11- multilingual12library_name: sentence-transformers13pipeline_tag: feature-extraction14---15 16# Granite Embedding 107M Multilingual17 18This is a copy of the [ibm-granite/granite-embedding-107m-multilingual](https://huggingface.co/ibm-granite/granite-embedding-107m-multilingual) model for document encoding purposes.19 20## Model Summary21Granite-Embedding-107M-Multilingual is a 107M parameter dense biencoder embedding model from the Granite Embeddings suite that can be used to generate high quality text embeddings. This model produces embedding vectors of size 384.22 23## Supported Languages24English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese.25 26## Usage27 28### With Sentence Transformers29```python30from sentence_transformers import SentenceTransformer31 32model = SentenceTransformer('RikoteMaster/MNLP_M3_document_encoder')33embeddings = model.encode(['Your text here'])34```35 36### With Transformers37```python38from transformers import AutoModel, AutoTokenizer39import torch40 41model = AutoModel.from_pretrained('RikoteMaster/MNLP_M3_document_encoder')42tokenizer = AutoTokenizer.from_pretrained('RikoteMaster/MNLP_M3_document_encoder')43 44inputs = tokenizer(['Your text here'], return_tensors='pt', padding=True, truncation=True)45with torch.no_grad():46    outputs = model(**inputs)47    embeddings = outputs.last_hidden_state[:, 0]  # CLS pooling48    embeddings = torch.nn.functional.normalize(embeddings, dim=1)49```50 51## Original Model52This model is based on [ibm-granite/granite-embedding-107m-multilingual](https://huggingface.co/ibm-granite/granite-embedding-107m-multilingual) by IBM.53