很多小伙伴可能會對pytorch怎么計(jì)算kl散度有些疑問,因?yàn)槭褂胮ytorch的函數(shù)算出來的結(jié)果與目標(biāo)值有一定差距,那么為什么會這樣呢?小編帶來了pytorch官方文檔,我們來看看官方文檔是怎么說的吧!
先附上官方文檔說明:https://pytorch.org/docs/stable/nn.functional.html
torch.nn.functional.kl_div(input, target, size_average=None, reduce=None, reduction='mean')
Parameters
input – Tensor of arbitrary shape
target – Tensor of the same shape as input
size_average (bool, optional) – Deprecated (see reduction). By default, the losses are averaged over each loss element in the batch. Note that for some losses, there multiple elements per sample. If the field size_average is set to False, the losses are instead summed for each minibatch. Ignored when reduce is False. Default: True
reduce (bool, optional) – Deprecated (see reduction). By default, the losses are averaged or summed over observations for each minibatch depending on size_average. When reduce is False, returns a loss per batch element instead and ignores size_average. Default: True
reduction (string, optional) – Specifies the reduction to apply to the output: 'none' | 'batchmean' | 'sum' | 'mean'. 'none': no reduction will be applied 'batchmean': the sum of the output will be divided by the batchsize 'sum': the output will be summed 'mean': the output will be divided by the number of elements in the output Default: 'mean'
然后看看怎么用:
第一個(gè)參數(shù)傳入的是一個(gè)對數(shù)概率矩陣,第二個(gè)參數(shù)傳入的是概率矩陣。這里很重要,不然求出來的kl散度可能是個(gè)負(fù)值。
比如現(xiàn)在我有兩個(gè)矩陣X, Y。因?yàn)閗l散度具有不對稱性,存在一個(gè)指導(dǎo)和被指導(dǎo)的關(guān)系,因此這連個(gè)矩陣輸入的順序需要確定一下。
舉個(gè)例子:
如果現(xiàn)在想用Y指導(dǎo)X,第一個(gè)參數(shù)要傳X,第二個(gè)要傳Y。就是被指導(dǎo)的放在前面,然后求相應(yīng)的概率和對數(shù)概率就可以了。
import torch
import torch.nn.functional as F
# 定義兩個(gè)矩陣
x = torch.randn((4, 5))
y = torch.randn((4, 5))
# 因?yàn)橐脃指導(dǎo)x,所以求x的對數(shù)概率,y的概率
logp_x = F.log_softmax(x, dim=-1)
p_y = F.softmax(y, dim=-1)
kl_sum = F.kl_div(logp_x, p_y, reduction='sum')
kl_mean = F.kl_div(logp_x, p_y, reduction='mean')
print(kl_sum, kl_mean)
>>> tensor(3.4165) tensor(0.1708)
補(bǔ)充:pytorch中的kl散度,為什么kl散度是負(fù)數(shù)?
F.kl_div()或者nn.KLDivLoss()是pytroch中計(jì)算kl散度的函數(shù),它的用法有很多需要注意的細(xì)節(jié)。
輸入
第一個(gè)參數(shù)傳入的是一個(gè)對數(shù)概率矩陣,第二個(gè)參數(shù)傳入的是概率矩陣。并且因?yàn)閗l散度具有不對稱性,存在一個(gè)指導(dǎo)和被指導(dǎo)的關(guān)系,因此這連個(gè)矩陣輸入的順序需要確定一下。如果現(xiàn)在想用Y指導(dǎo)X,第一個(gè)參數(shù)要傳X,第二個(gè)要傳Y。就是被指導(dǎo)的放在前面,然后求相應(yīng)的概率和對數(shù)概率就可以了。
所以,一隨機(jī)初始化一個(gè)tensor為例,對于第一個(gè)輸入,我們需要先對這個(gè)tensor進(jìn)行softmax(確保各維度和為1),然后再取log;對于第二個(gè)輸入,我們需要對這個(gè)tensor進(jìn)行softmax。
import torch
import torch.nn.functional as F
a = torch.tensor([[0,0,1.1,2,0,10,0],[0,0,1,2,0,10,0]])
log_a =F.log_softmax(a)
b = torch.tensor([[0,0,1.1,2,0,7,0],[0,0,1,2,0,10,0]])
softmax_b =F.softmax(b,dim=-1)
kl_mean = F.kl_div(log_a, softmax_b, reduction='mean')
print(kl_mean)
為什么KL散度計(jì)算出來為負(fù)數(shù)
先確保對第一個(gè)輸入進(jìn)行了softmax+log操作,對第二個(gè)參數(shù)進(jìn)行了softmax操作。不進(jìn)行softmax操作就可能為負(fù)。
然后查看自己的輸入是否是小數(shù)點(diǎn)后有很多位,當(dāng)小數(shù)點(diǎn)后很多位的時(shí)候,pytorch下的softmax會產(chǎn)生各維度和不為1的現(xiàn)象,導(dǎo)致kl散度為負(fù),如下所示:
a = torch.tensor([[0.,0,0.000001,0.0000002,0,0.0000007,0]])
log_a =F.log_softmax(a,dim=-1)
print("log_a:",log_a)
b = torch.tensor([[0.,0,0.000001,0.0000002,0,0.0000007,0]])
softmax_b =F.softmax(b,dim=-1)
print("softmax_b:",softmax_b)
kl_mean = F.kl_div(log_a, softmax_b,reduction='mean')
print("kl_mean:",kl_mean)
輸出如下,我們可以看到softmax_b的各維度和不為1:
以上就是pytorch怎么計(jì)算kl散度的全部內(nèi)容,希望能給大家一個(gè)參考,也希望大家多多支持W3Cschool。