""" The following code has been inspired from https://huggingface.co/sentence-transformers/all-mpnet-base-v2 as an example to explain the concept of semantic search using ANN """ from transformers import AutoTokenizer, AutoModel import torch import torch.nn.functional as F class SemanticEmbedding: def __init__(self, model_name='sentence-transformers/all-mpnet-base-v2'): self.tokenizer = AutoTokenizer.from_pretrained(model_name) self.model = AutoModel.from_pretrained(model_name) #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(self, model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) def get_embedding(self, sentences): # Tokenize sentences encoded_input = self.tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') with torch.no_grad(): model_output = self.model(**encoded_input) # Perform pooling sentence_embeddings = self.mean_pooling(model_output, encoded_input['attention_mask']) # Normalize embeddings sentence_embeddings = F.normalize(sentence_embeddings, p=2, dim=1) return sentence_embeddings.detach().numpy()