import os
import paddle
from paddlenlp.transformers import AutoModel, AutoTokenizer
import paddle.nn as nn
import paddle.nn.functional as F
import numpy as np
from paddle import inference
from tqdm import tqdm
from paddlenlp.data import Pad, Tuple
class SimCSE(nn.Layer):
def __init__(self, pretrained_model, dropout=None, margin=0.0, scale=20, output_emb_size=None):
super().__init__()
self.ptm = pretrained_model#预训练模型
#dropout is not None和dropout是不一样的,dropout＝０．时,dropout是Ｆalse,dropout is not None是Ｔrue
self.dropout = nn.Dropout(dropout if dropout is not None else 0.1)
self.output_emb_size = output_emb_size
if output_emb_size > 0:#如果output_emb_size>0,线性转换
weight_attr = paddle.ParamAttr(initializer=paddle.nn.initializer.TruncatedNormal(std=0.02))
self.emb_reduce_linear = paddle.nn.Linear(768, output_emb_size, weight_attr=weight_attr)
self.margin = margin
self.scale = scale
@paddle.jit.to_static(
input_spec=[
paddle.static.InputSpec(shape=[None, None], dtype="int64"),
paddle.static.InputSpec(shape=[None, None], dtype="int64"),
]
)
def get_pooled_embedding(
self, input_ids, token_type_ids=None, position_ids=None, attention_mask=None, with_pooler=True
):
# Note: cls_embedding is poolerd embedding with act tanh
sequence_output, cls_embedding = self.ptm(input_ids, token_type_ids, position_ids, attention_mask)
if with_pooler is False:#如果ptm不返回池化层,把［CLS］输出作为池化输出
cls_embedding = sequence_output[:, 0, :]
if self.output_emb_size > 0:
cls_embedding = self.emb_reduce_linear(cls_embedding)
cls_embedding = self.dropout(cls_embedding)
cls_embedding = F.normalize(cls_embedding, p=2, axis=-1)#向量单位化（b,d）
return cls_embedding
def get_semantic_embedding(self, data_loader):
self.eval()
with paddle.no_grad():
for batch_data in data_loader:
input_ids, token_type_ids = batch_data
text_