RRFRRF
init commit without .pth
dee113c
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
import torch
import torch.nn as nn
import torch
from torch.autograd import Variable
import copy
import torch.nn.functional as F
from torch.nn import CrossEntropyLoss, MSELoss
class Model(nn.Module):
def __init__(self, encoder,config,tokenizer,args):
super(Model, self).__init__()
self.encoder = encoder
self.config=config
self.tokenizer=tokenizer
self.args=args
def forward(self, input_ids=None,p_input_ids=None,n_input_ids=None,labels=None):
bs,_=input_ids.size()
input_ids=torch.cat((input_ids,p_input_ids,n_input_ids),0)
outputs=self.encoder(input_ids,attention_mask=input_ids.ne(1))
if len(outputs) > 1:
outputs = outputs[1]
else:
outputs = outputs[0][:, 0, :]
outputs=outputs.split(bs,0)
prob_1=(outputs[0]*outputs[1]).sum(-1)
prob_2=(outputs[0]*outputs[2]).sum(-1)
temp=torch.cat((outputs[0],outputs[1]),0)
temp_labels=torch.cat((labels,labels),0)
prob_3= torch.mm(outputs[0],temp.t())
mask=labels[:,None]==temp_labels[None,:]
prob_3=prob_3*(1-mask.float())-1e9*mask.float()
prob=torch.softmax(torch.cat((prob_1[:,None],prob_2[:,None],prob_3),-1),-1)
loss=torch.log(prob[:,0]+1e-10)
loss=-loss.mean()
return loss,outputs[0]