Online hard example mining PyTorch

NLL_OHEM.py

import torch as th                                                                 
                                                                                   
                                                                                   
class NLL_OHEM(th.nn.NLLLoss):                                                     
    """ Online hard example mining. 
    Needs input from nn.LogSotmax() """                                             
                                                                                   
    def __init__(self, ratio):      
        super(NLL_OHEM, self).__init__(None, True)                                 
        self.ratio = ratio                                                         
                                                                                   
    def forward(self, x, y, ratio=None):                                           
        if ratio is not None:                                                      
            self.ratio = ratio                                                     
        num_inst = x.size(0)                                                       
        num_hns = int(self.ratio * num_inst)                                       
        x_ = x.clone()                                                             
        inst_losses = th.autograd.Variable(th.zeros(num_inst)).cuda()              
        for idx, label in enumerate(y.data):                                       
            inst_losses[idx] = -x_.data[idx, label]                                 
        #loss_incs = -x_.sum(1)                                                    
        _, idxs = inst_losses.topk(num_hns)                                        
        x_hn = x.index_select(0, idxs)                                             
        y_hn = y.index_select(0, idxs)                                             
        return th.nn.functional.nll_loss(x_hn, y_hn)