必須將權(quán)重也轉(zhuǎn)為Tensor的cuda格式；
將該class_weight作為交叉熵函數(shù)對(duì)應(yīng)參數(shù)的輸入值。
class_weight = torch.FloatTensor([0.13859937, 0.5821059, 0.63871904, 2.30220396, 7.1588294, 0]).cuda()
補(bǔ)充：關(guān)于pytorch的CrossEntropyLoss的weight參數(shù)
你可以試試下面代碼
import torch import torch.nn as nn inputs = torch.FloatTensor([0,1,0,0,0,1]) outputs = torch.LongTensor([0,1]) inputs = inputs.view((1,3,2)) outputs = outputs.view((1,2)) weight_CE = torch.FloatTensor([1,1,1]) ce = nn.CrossEntropyLoss(ignore_index=255,weight=weight_CE) loss = ce(inputs,outputs) print(loss)
tensor(1.4803)
這里的手動(dòng)計(jì)算是：
loss1 = 0 + ln(e0 + e0 + e0) = 1.098
loss2 = 0 + ln(e1 + e0 + e1) = 1.86
求平均 = (loss1 *1 + loss2 *1)/ 2 = 1.4803
import torch import torch.nn as nn inputs = torch.FloatTensor([0,1,0,0,0,1]) outputs = torch.LongTensor([0,1]) inputs = inputs.view((1,3,2)) outputs = outputs.view((1,2)) weight_CE = torch.FloatTensor([1,2,3]) ce = nn.CrossEntropyLoss(ignore_index=255,weight=weight_CE) loss = ce(inputs,outputs) print(loss)
tensor(1.6075)
手算發(fā)現(xiàn)，并不是單純的那權(quán)重相乘：
loss1 = 0 + ln(e0 + e0 + e0) = 1.098
loss2 = 0 + ln(e1 + e0 + e1) = 1.86
求平均 = (loss1 * 1 + loss2 * 2)/ 2 = 2.4113
而是
loss1 = 0 + ln(e0 + e0 + e0) = 1.098
loss2 = 0 + ln(e1 + e0 + e1) = 1.86
求平均 = (loss1 *1 + loss2 *2) / 3 = 1.6075
發(fā)現(xiàn)了么，加權(quán)后，除以的是權(quán)重的和，不是數(shù)目的和。
import torch import torch.nn as nn inputs = torch.FloatTensor([0,1,2,0,0,0,0,0,0,1,0,0.5]) outputs = torch.LongTensor([0,1,2,2]) inputs = inputs.view((1,3,4)) outputs = outputs.view((1,4)) weight_CE = torch.FloatTensor([1,2,3]) ce = nn.CrossEntropyLoss(weight=weight_CE) # ce = nn.CrossEntropyLoss(ignore_index=255) loss = ce(inputs,outputs) print(loss)
tensor(1.5472)
手算：
loss1 = 0 + ln(e0 + e0 + e0) = 1.098
loss2 = 0 + ln(e1 + e0 + e1) = 1.86
loss3 = 0 + ln(e2 + e0 + e0) = 2.2395
loss4 = -0.5 + ln(e0.5 + e0 + e0) = 0.7943
求平均 = (loss1 * 1 + loss2 * 2+loss3 * 3+loss4 * 3) / 9 = 1.5472
可能有人對(duì)loss的CE計(jì)算過(guò)程有疑問(wèn)，我這里細(xì)致寫寫交叉熵的計(jì)算過(guò)程，就拿最后一個(gè)例子的loss4的計(jì)算說(shuō)明
以上為個(gè)人經(jīng)驗(yàn)，希望能給大家一個(gè)參考，也希望大家多多支持腳本之家。
標(biāo)簽：成都 常州 山東 宿遷 駐馬店 六盤水 江蘇 蘭州
巨人網(wǎng)絡(luò)通訊聲明：本文標(biāo)題《pytorch交叉熵?fù)p失函數(shù)的weight參數(shù)的使用》，本文關(guān)鍵詞 pytorch,交叉,熵,損失,函數(shù),；如發(fā)現(xiàn)本文內(nèi)容存在版權(quán)問(wèn)題，煩請(qǐng)?zhí)峁┫嚓P(guān)信息告之我們，我們將及時(shí)溝通與處理。本站內(nèi)容系統(tǒng)采集于網(wǎng)絡(luò)，涉及言論、版權(quán)與本站無(wú)關(guān)。
