PyTorch遇到的error2

廖星宇 多项式拟合

UserWarning: invalid index of a 0-dim tensor. This will be an error in PyTorch 0.5. Use tensor.item() to convert a 0-dim tensor to a Python number 

将pytorch更新到0.4.0最新版后对0.3.1版本代码会有如下警告，它在提醒用户下个版本这将成为一个错误

修改：

#    print_loss = loss.data[0]
    print_loss = loss.item()

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
"""
Created on Fri Sep 14 16:19:59 2018

@author: lthpc
"""

import numpy as np
import matplotlib.pyplot as plt
import torch
from torch.autograd import Variable
from torch import nn,optim

def make_features(x):
    '''Builds features i.e. a matrix with columns [x, x^2, x^3].'''
    x = x.unsqueeze(1)
    return torch.cat([x ** i for i in range(1,4)],1)

W_target = torch.FloatTensor([0.5,3,2.4]).unsqueeze(1)
b_target = torch.FloatTensor([0.9])

def f(x):
    '''Approximated function.'''
    return x.mm(W_target)+b_target[0]

def get_batch(batch_size=32):
    '''Builds a batch i.e. (x,f(x)) pair.'''
    random = torch.randn(batch_size)
    x = make_features(random)
    y = f(x)
    if torch.cuda.is_available():
        return Variable(x).cuda(), Variable(y).cuda()
    else:
        return Variable(x), Variable(y)
    
#    Define model
class poly_model(nn.Module):
    def __init__(self):
        super(poly_model,self).__init__()
        self.poly = nn.Linear(3,1)
        
    def forward(self,x):
        out = self.poly(x)
        return out
    
if torch.cuda.is_available():
    model = poly_model().cuda()
else:
    model = poly_model()

criterion = nn.MSELoss()
optimizer = optim.SGD(model.parameters(),lr=1e-3)

epoch = 0
while True:
    #Get data
    batch_x,batch_y = get_batch()
    # Forward pass
    output = model(batch_x)
    loss = criterion(output,batch_y)
#    print_loss = loss.data[0]
    print_loss = loss.item()
    #Reset gradients
    optimizer.zero_grad()
    #Backward pass
    loss.backward()
    print('loss:%.5f\n'%loss)
        #update parameters
    optimizer.step()
    epoch+=1
    if print_loss<1e-3:
        break

x = np.linspace(-1,1,30)
x_sample = torch.from_numpy(x)
x_sample = x_sample.unsqueeze(1)
x_sample = torch.cat([x_sample ** i for i in range(1,4)] , 1)
x_sample = x_sample.float()
y_actural = f(x_sample)
tt = x_sample.cuda()
y_predict = model(tt)
plt.plot(x,y_actural.numpy(),'ro',x,y_predict.data.cpu().numpy())
plt.legend(['real point','fit'])
plt.show()