问题描述

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [256, 1]], which is output 0 of TBackward, is at version 2; expected version 1 instead. Hint: the backtrace further above shows the operation that failed to compute its gradient. The variable in question was changed in there or anywhere later. Good luck!

 这个问题在于使用了更新之前的变量计算loss，但是这些变量出现了更新，导致梯度反传时找不到用于计算loss的变量，进而报错。

 解决方案

这个问题很好的一个示例是生成对抗网络生成器和判别器的训练过程：

解决参考链接：

https://github.com/pytorch/pytorch/issues/39141

 源代码是这样的

import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt


# Hyper Parameters
BATCH_SIZE = 64
LR_G = 0.0001
LR_D = 0.0001 
N_IDEAS = 5  
ART_COMPONENTS = 15 
PAINT_POINTS = np.vstack([np.linspace(-1, 1, ART_COMPONENTS) for _ in range(BATCH_SIZE)])


def artist_works():  # painting from the famous artist (real target)
    r = 0.02 * np.random.randn(1, ART_COMPONENTS)
    paintings = np.sin(PAINT_POINTS * np.pi) + r
    paintings = torch.from_numpy(paintings).float()
    return paintings


G = nn.Sequential(  # Generator
    nn.Linear(N_IDEAS, 128),  # random ideas (could from normal distribution)
    nn.ReLU(),
    nn.Linear(128, ART_COMPONENTS),  # making a painting from these random ideas
)

D = nn.Sequential(  # Discriminator
    nn.Linear(ART_COMPONENTS, 128),  # receive art work either from the famous artist or a newbie like G
    nn.ReLU(),
    nn.Linear(128, 1),
    nn.Sigmoid(),  # tell the probability that the art work is made by artist
)

opt_D = torch.optim.Adam(D.parameters(), lr=LR_D)
opt_G = torch.optim.Adam(G.parameters(), lr=LR_G)


for step in range(10000):
    artist_paintings = artist_works()  # real painting from artist
    G_ideas = torch.randn(BATCH_SIZE, N_IDEAS)  # random ideas
    G_paintings = G(G_ideas)  # fake painting from G (random ideas)

    prob_artist0 = D(artist_paintings)  # D try to increase this prob
    prob_artist1 = D(G_paintings)  # D try to reduce this prob

    D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1. - prob_artist1))
    G_loss = torch.mean(torch.log(1. - prob_artist1))

    opt_D.zero_grad()
    D_loss.backward(retain_graph=True)  # reusing computational graph
    opt_D.step()

    opt_G.zero_grad()
    G_loss.backward()
    opt_G.step()

解决思路有两条：第一，将python版本降至1.4以下（含），第二，将网络的更新部分写成以下形式：

for step in range(10000):
    artist_paintings = artist_works()  # real painting from artist
    G_ideas = torch.randn(BATCH_SIZE, N_IDEAS)  # random ideas
    G_paintings = G(G_ideas)  # fake painting from G (random ideas)

    prob_artist1 = D(G_paintings)  # G tries to fool D

    G_loss = torch.mean(torch.log(1. - prob_artist1))
    opt_G.zero_grad()
    G_loss.backward()
    opt_G.step()

    prob_artist0 = D(artist_paintings)  # D try to increase this prob
    # detach here to make sure we don't backprop in G that was already changed.
    prob_artist1 = D(G_paintings.detach())  # D try to reduce this prob

    D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1. - prob_artist1))
    opt_D.zero_grad()
    D_loss.backward(retain_graph=True)  # reusing computational graph
    opt_D.step()

即在生成器更新完之后使用更新之后的生成器输出结果来计算判别器loss，用新计算出来的loss对判别器进行更新，这样的话就可以解决问题描述中出现的Bug。

参考链接

关于GAN训练过程中的报错：one of the variables needed for gradient computation has been modified by an inplace_训练w-gan时one of the variables needed for gradient c-CSDN博客