sentence_transformers 教程_sentence transformer loss

文档：Search — Sentence-Transformers documentation

用途：

该模主要用来做句子嵌入，下游常用来做语意匹配

ContrastiveLoss

别名：

pairwise ranking loss

pairwise loss

公式:

 loss = y*||anchor - positive|| +(1-y)*max(margin-||anchor - negative||, 0).

# 损失函数代码表示，摘自 sentence transformers
losses = 0.5 * (labels.float() * distances.pow(2) + (1 - labels).float() * F.relu(self.margin - distances).pow(2))

from sentence_transformers import SentenceTransformer, LoggingHandler, losses, InputExample
from torch.utils.data import DataLoader
 
model = SentenceTransformer('all-MiniLM-L6-v2')
train_examples = [
    InputExample(texts=['This is a positive pair', 'Where the distance will be minimized'], label=1),
    InputExample(texts=['This is a negative pair', 'Their distance will be increased'], label=0)]
 
train_dataloader = DataLoader(train_examples, shuffle=True, batch_size=2)
train_loss = losses.ContrastiveLoss(model=model)
 
model.fit([(train_dataloader, train_loss)], show_progress_bar=True)

CosineSimilarityLoss

计算出样本的余弦相似度，和label做MSE损失

from sentence_transformers import SentenceTransformer, InputExample, losses
from torch.utils.data import DataLoader
 
#Define the model. Either from scratch of by loading a pre-trained model
model = SentenceTransformer('distilbert-base-nli-mean-tokens')
 
#Define your train examples. You need more than just two examples...
train_examples = [InputExample(texts=['My first sentence', 'My second sentence'], label=0.8),
    InputExample(texts=['Another pair', 'Unrelated sentence'], label=0.3)]
 
#Define your train dataset, the dataloader and the train loss
train_dataloader = DataLoader(train_examples, shuffle=True, batch_size=16)
train_loss = losses.CosineSimilarityLoss(model)
 
#Tune the model
model.fit(train_objectives=[(train_dataloader, train_loss)], epochs=1, warmup_steps=100)

MultipleNegativesRankingLoss

对比损失，同一批次的，其它样本视为负样本，分别两两求余弦相似度，最后做交叉熵损失，正样本的得分应该最高

train_examples = [InputExample(texts=['Anchor 1', 'Positive 1']),
                InputExample(texts=['Anchor 2', 'Positive 2'])]
train_dataloader = DataLoader(train_examples, shuffle=True, batch_size=32)
train_loss = losses.MultipleNegativesRankingLoss(model=model)

TripletLoss

loss = max(||anchor - positive|| - ||anchor - negative|| + margin, 0).

from sentence_transformers import SentenceTransformer,  SentencesDataset, LoggingHandler, losses
from sentence_transformers.readers import InputExample
 
model = SentenceTransformer('distilbert-base-nli-mean-tokens')
train_examples = [InputExample(texts=['Anchor 1', 'Positive 1', 'Negative 1']),
    InputExample(texts=['Anchor 2', 'Positive 2', 'Negative 2'])]
train_dataset = SentencesDataset(train_examples, model)
train_dataloader = DataLoader(train_dataset, shuffle=True, batch_size=train_batch_size)
train_loss = losses.TripletLoss(model=model)