本文主要是介绍pycharm安装pytorch报错 提示系列问题 torch 包找不到因为pip版本低，结果升级了pip从19.3到20.2 4又提示高版不支持torch安装，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！
pycharm安装pytorch报错 提示系列问题 torch 包找不到因为pip版本低，结果升级了pip从19.3到20.2 4又提示高版不支持torch安装
DEPRECATION: The -b/--build/--build-dir/--build-directory option is deprecated. pip 20.3 will remove support for this functionality. A possible replacement is use the TMPDIR/TEMP/TMP environment variable, possibly combined with --no-clean. You can find discussion regarding this at https://github.com/pypa/pip/issues/8333.
如上报错来自参考
https://www.jb51.net/article/194349.htm
参考有同学遇到类似的问题的解决方案
https://blog.csdn.net/weixin_41010198/article/details/103107083
ERROR: Could not find a version that satisfies the requirement torch==1.6.0 (from versions: 0.1.2, 0.1.2.post1, 0.1.2.post2)
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
DEPRECATION: The -b/--build/--build-dir/--build-directory option is deprecated. pip 20.3 will remove support for this functionality. A possible replacement is use the TMPDIR/TEMP/TMP environment variable, possibly combined with --no-clean. You can find discussion regarding this at https://github.com/pypa/pip/issues/8333.
ERROR: Could not find a version that satisfies the requirement torch==1.6.0 (from versions: 0.1.2, 0.1.2.post1, 0.1.2.post2)
ERROR: No matching distribution found for torch==1.6.0
三 最终解决方案
是参考的如下的方式解决pycharm 导入pytorch问题的，真是周折
主要参考的如下连接的，将pycharm的system interceptor切换为anaconda的python.exe的环境变量
https://www.jb51.net/article/181954.htm
同时从上面的博文中也可以得出切换pycharm的环境变量的方法，可以使在pycharm中进行如下设置
之所以采用这种方案，是因为代码在本地启动的anaconda jupyter notebook中是可以调用pytorch的，版本是1.1.0，但是在pycharm中就是安装不上，无论是1.1.0 还是0.41 1.6的版本都安装不上，分析可能是环境变量问题，
萌生了使用anaconda的环境变量 （主要是编译器）的想法，看到如上博文，参考更改了项目的解析器，代码中引用的torch包就找到了，不报找不到对应包的问题了
实验代码为NLP 的Seq2seq代码
# code by Tae Hwan Jung(Jeff Jung) @graykode, modify by wmathor
import torch
import numpy as np
import torch.nn as nn
import torch.utils.data as Datadevice = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# S: Symbol that shows starting of decoding input
# E: Symbol that shows starting of decoding output
# ?: Symbol that will fill in blank sequence if current batch data size is short than n_stepletter = [c for c in 'SE?abcdefghijklmnopqrstuvwxyz']
letter2idx = {n: i for i, n in enumerate(letter)}seq_data = [['man', 'women'], ['black', 'white'], ['king', 'queen'], ['girl', 'boy'], ['up', 'down'], ['high', 'low']]# Seq2Seq Parameter
n_step = max([max(len(i), len(j)) for i, j in seq_data]) # max_len(=5)
n_hidden = 128
n_class = len(letter2idx) # classfication problem
batch_size = 3#3
def make_data(seq_data):enc_input_all, dec_input_all, dec_output_all = [], [], []for seq in seq_data:for i in range(2):seq[i] = seq[i] + '?' * (n_step - len(seq[i])) # 'man??', 'women'enc_input = [letter2idx[n] for n in (seq[0] + 'E')] # ['m', 'a', 'n', '?', '?', 'E']dec_input = [letter2idx[n] for n in ('S' + seq[1])] # ['S', 'w', 'o', 'm', 'e', 'n']dec_output = [letter2idx[n] for n in (seq[1] + 'E')] # ['w', 'o', 'm', 'e', 'n', 'E']enc_input_all.append(np.eye(n_class)[enc_input])dec_input_all.append(np.eye(n_class)[dec_input])dec_output_all.append(dec_output) # not one-hot# make tensorreturn torch.Tensor(enc_input_all), torch.Tensor(dec_input_all), torch.LongTensor(dec_output_all)'''
enc_input_all: [6, n_step+1 (because of 'E'), n_class]
dec_input_all: [6, n_step+1 (because of 'S'), n_class]
dec_output_all: [6, n_step+1 (because of 'E')]
'''
enc_input_all, dec_input_all, dec_output_all = make_data(seq_data)
#4
class TranslateDataSet(Data.Dataset):def __init__(self, enc_input_all, dec_input_all, dec_output_all):self.enc_input_all = enc_input_allself.dec_input_all = dec_input_allself.dec_output_all = dec_output_alldef __len__(self): # return dataset sizereturn len(self.enc_input_all)def __getitem__(self, idx):return self.enc_input_all[idx], self.dec_input_all[idx], self.dec_output_all[idx]loader = Data.DataLoader(TranslateDataSet(enc_input_all, dec_input_all, dec_output_all), batch_size, True)
#5
def make_data(seq_data):enc_input_all, dec_input_all, dec_output_all = [], [], []for seq in seq_data:for i in range(2):seq[i] = seq[i] + '?' * (n_step - len(seq[i])) # 'man??', 'women'enc_input = [letter2idx[n] for n in (seq[0] + 'E')] # ['m', 'a', 'n', '?', '?', 'E']dec_input = [letter2idx[n] for n in ('S' + seq[1])] # ['S', 'w', 'o', 'm', 'e', 'n']dec_output = [letter2idx[n] for n in (seq[1] + 'E')] # ['w', 'o', 'm', 'e', 'n', 'E']enc_input_all.append(np.eye(n_class)[enc_input])dec_input_all.append(np.eye(n_class)[dec_input])dec_output_all.append(dec_output) # not one-hot# make tensorreturn torch.Tensor(enc_input_all), torch.Tensor(dec_input_all), torch.LongTensor(dec_output_all)'''
enc_input_all: [6, n_step+1 (because of 'E'), n_class]
dec_input_all: [6, n_step+1 (because of 'S'), n_class]
dec_output_all: [6, n_step+1 (because of 'E')]
'''
enc_input_all, dec_input_all, dec_output_all = make_data(seq_data)
#6
class TranslateDataSet(Data.Dataset):def __init__(self, enc_input_all, dec_input_all, dec_output_all):self.enc_input_all = enc_input_allself.dec_input_all = dec_input_allself.dec_output_all = dec_output_alldef __len__(self): # return dataset sizereturn len(self.enc_input_all)def __getitem__(self, idx):return self.enc_input_all[idx], self.dec_input_all[idx], self.dec_output_all[idx]loader = Data.DataLoader(TranslateDataSet(enc_input_all, dec_input_all, dec_output_all), batch_size, True)
#7# Model
class Seq2Seq(nn.Module):def __init__(self):super(Seq2Seq, self).__init__()self.encoder = nn.RNN(input_size=n_class, hidden_size=n_hidden, dropout=0.5) # encoderself.decoder = nn.RNN(input_size=n_class, hidden_size=n_hidden, dropout=0.5) # decoderself.fc = nn.Linear(n_hidden, n_class)def forward(self, enc_input, enc_hidden, dec_input):# enc_input(=input_batch): [batch_size, n_step+1, n_class]# dec_inpu(=output_batch): [batch_size, n_step+1, n_class]enc_input = enc_input.transpose(0, 1) # enc_input: [n_step+1, batch_size, n_class]dec_input = dec_input.transpose(0, 1) # dec_input: [n_step+1, batch_size, n_class]# h_t : [num_layers(=1) * num_directions(=1), batch_size, n_hidden]_, h_t = self.encoder(enc_input, enc_hidden)# outputs : [n_step+1, batch_size, num_directions(=1) * n_hidden(=128)]outputs, _ = self.decoder(dec_input, h_t)model = self.fc(outputs) # model : [n_step+1, batch_size, n_class]return modelmodel = Seq2Seq().to(device)
criterion = nn.CrossEntropyLoss().to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
#8
for epoch in range(5000):for enc_input_batch, dec_input_batch, dec_output_batch in loader:# make hidden shape [num_layers * num_directions, batch_size, n_hidden]h_0 = torch.zeros(1, batch_size, n_hidden).to(device)(enc_input_batch, dec_intput_batch, dec_output_batch) = (enc_input_batch.to(device), dec_input_batch.to(device), dec_output_batch.to(device))# enc_input_batch : [batch_size, n_step+1, n_class]# dec_intput_batch : [batch_size, n_step+1, n_class]# dec_output_batch : [batch_size, n_step+1], not one-hotpred = model(enc_input_batch, h_0, dec_intput_batch)# pred : [n_step+1, batch_size, n_class]pred = pred.transpose(0, 1) # [batch_size, n_step+1(=6), n_class]loss = 0for i in range(len(dec_output_batch)):# pred[i] : [n_step+1, n_class]# dec_output_batch[i] : [n_step+1]loss += criterion(pred[i], dec_output_batch[i])if (epoch + 1) % 1000 == 0:print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.6f}'.format(loss))optimizer.zero_grad()loss.backward()optimizer.step()
#9
# Test
def translate(word):enc_input, dec_input, _ = make_data([[word, '?' * n_step]])enc_input, dec_input = enc_input.to(device), dec_input.to(device)# make hidden shape [num_layers * num_directions, batch_size, n_hidden]print("enc_input=",enc_input, "dec_input=",dec_input,"n_hidden=",n_hidden)hidden = torch.zeros(1, 1, n_hidden).to(device)output = model(enc_input, hidden, dec_input)# output : [n_step+1, batch_size, n_class]print("output.data=",output.data)predict = output.data.max(2, keepdim=True)[1] # select n_class dimensionprint("predict=",predict)decoded = [letter[i] for i in predict]translated = ''.join(decoded[:decoded.index('E')])return translated.replace('?', '')print('test')
print('man ->', translate('man'))
print('mans ->', translate('mans'))
print('king ->', translate('king'))
print('black ->', translate('black'))
print('up ->', translate('up'))
#10
#执行结果C:\ProgramData\Anaconda3\python.exe C:/Users/pc/PycharmProjects/seq2seq/Seq2Seq.py
C:\ProgramData\Anaconda3\lib\site-packages\torch\nn\modules\rnn.py:60: UserWarning: dropout option adds dropout after all but last recurrent layer, so non-zero dropout expects num_layers greater than 1, but got dropout=0.5 and num_layers=1
"num_layers={}".format(dropout, num_layers))
Epoch: 1000 cost = 0.001993
Epoch: 1000 cost = 0.002054
Epoch: 2000 cost = 0.000441
Epoch: 2000 cost = 0.000415
Epoch: 3000 cost = 0.000128
Epoch: 3000 cost = 0.000134
Epoch: 4000 cost = 0.000043
Epoch: 4000 cost = 0.000045
Epoch: 5000 cost = 0.000015
Epoch: 5000 cost = 0.000016
test
enc_input= tensor([[[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) dec_input= tensor([[[1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) n_hidden= 128
output.data= tensor([[[-4.3065e+00, -4.8961e+00, -1.0978e+00, -4.1016e+00, 1.6963e+00,
-3.4909e+00, 1.3702e+00, -3.3739e-02, -4.2637e+00, -3.8116e+00,
-1.6880e+00, -2.9487e+00, -4.1300e+00, -3.1433e+00, 1.6024e+00,
-2.8452e+00, 1.0767e+00, -5.6262e+00, -4.0048e+00, 1.0652e+00,
-4.4673e+00, -4.5837e+00, -4.2722e+00, -3.1410e+00, -3.9433e+00,
1.5158e+01, -4.0215e+00, -4.3383e+00, -4.2129e+00]],
[[-2.3903e+00, 7.6515e-01, 2.1158e+00, -2.1588e+00, -3.0337e+00,
-2.6551e+00, 7.7586e-01, -5.4225e+00, -2.8381e+00, -2.8219e+00,
2.1630e+00, -2.2365e+00, -3.1426e+00, -2.7296e+00, 1.1102e+00,
-7.0369e-01, 2.7494e+00, 1.6497e+01, -2.7483e+00, 4.9587e-01,
-2.6725e+00, -3.1282e+00, -2.2622e+00, 1.7541e+00, -2.8997e+00,
-6.0975e+00, -2.9473e+00, -4.8188e+00, -2.5179e+00]],
[[-2.6203e+00, 1.7803e+00, 1.8365e+00, -2.0814e+00, -8.7015e-02,
-1.9189e+00, -4.1026e+00, 1.7874e+00, -1.9336e+00, -1.9122e+00,
-7.5097e+00, 1.6539e+00, -2.0918e+00, -2.1359e+00, -1.5856e+00,
1.5040e+01, -6.0545e+00, 1.0486e+00, -2.1752e+00, -3.8106e+00,
-2.2595e+00, -2.4848e+00, -1.9771e+00, 1.5679e+00, -2.1174e+00,
2.0894e+00, -1.6449e+00, 1.7625e+00, -2.0804e+00]],
[[-2.4143e+00, 2.9779e+00, 3.3485e-01, -1.7429e+00, -4.8116e+00,
-1.6712e+00, -1.8053e+00, 1.5846e+01, -2.0152e+00, -1.9535e+00,
-5.5402e+00, -7.0520e+00, -2.2360e+00, -1.9970e+00, -2.4646e+00,
-2.3482e+00, 2.8388e+00, -4.2379e+00, -2.1541e+00, -4.0036e-01,
-1.7277e+00, -2.3172e+00, 2.6084e+00, -2.3780e+00, -1.9208e+00,
2.9743e+00, -1.7947e+00, -2.6649e-02, -1.6980e+00]],
[[-1.4365e+00, 1.1394e+00, 3.4600e+00, -1.5666e+00, -1.6747e+00,
-1.5963e+00, -1.9778e-01, 2.6304e+00, -1.2731e+00, -1.4576e+00,
-3.7625e+00, -7.0933e+00, -1.2083e+00, -1.6387e+00, -4.2228e+00,
-3.8364e+00, 1.6177e+01, -4.5725e-01, -1.2973e+00, 6.8912e-01,
-1.9087e+00, -1.7284e+00, -6.8999e+00, 6.5842e-03, -1.5360e+00,
1.2516e+00, -1.2836e+00, -1.3752e+00, -1.4469e+00]],
[[-8.9884e-01, 1.8602e+01, 3.8394e+00, -7.6437e-01, -2.0698e+00,
-7.1664e-01, 3.8823e+00, -7.9843e-01, -1.1896e+00, -3.4626e-01,
-7.7051e+00, -7.4298e+00, -7.6236e-01, -9.4240e-01, -1.5845e+00,
3.8690e+00, 2.7345e+00, 2.6617e+00, -3.8022e-01, -1.0083e+00,
-8.5122e-01, -1.0864e+00, -2.5288e+00, 9.2861e-01, -9.4337e-01,
-3.5223e+00, -5.6646e-01, -6.5137e+00, -9.3167e-01]]])
predict= tensor([[[25]],
[[17]],
[[15]],
[[ 7]],
[[16]],
[[ 1]]])
man -> women
enc_input= tensor([[[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) dec_input= tensor([[[1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) n_hidden= 128
output.data= tensor([[[-4.2364e+00, -5.1443e+00, -1.2253e+00, -3.9696e+00, 1.7959e+00,
-3.4189e+00, 1.1670e+00, 1.4500e-01, -4.1254e+00, -3.6900e+00,
-1.5142e+00, -2.7320e+00, -4.0051e+00, -3.0514e+00, 1.4039e+00,
-2.6893e+00, 8.5148e-01, -5.8578e+00, -3.9052e+00, 8.6857e-01,
-4.3486e+00, -4.4292e+00, -4.1178e+00, -3.0776e+00, -3.8222e+00,
1.5356e+01, -3.9074e+00, -4.1002e+00, -4.0902e+00]],
[[-2.3948e+00, 5.5543e-01, 2.1681e+00, -2.1550e+00, -2.9457e+00,
-2.6404e+00, 7.8076e-01, -5.3944e+00, -2.8324e+00, -2.8036e+00,
2.5039e+00, -2.1335e+00, -3.1115e+00, -2.7270e+00, 1.1344e+00,
-8.8501e-01, 2.7650e+00, 1.6197e+01, -2.7496e+00, 5.4389e-01,
-2.6732e+00, -3.1151e+00, -2.1891e+00, 1.6821e+00, -2.8828e+00,
-6.0967e+00, -2.9272e+00, -4.7865e+00, -2.5106e+00]],
[[-2.6012e+00, 1.6449e+00, 1.9008e+00, -2.0793e+00, -6.8719e-02,
-1.8982e+00, -4.1096e+00, 1.7365e+00, -1.9221e+00, -1.9143e+00,
-7.3788e+00, 1.9041e+00, -2.0739e+00, -2.1198e+00, -1.5268e+00,
1.5047e+01, -6.1965e+00, 1.0572e+00, -2.1663e+00, -3.7946e+00,
-2.2383e+00, -2.4685e+00, -1.9318e+00, 1.5860e+00, -2.1032e+00,
1.9331e+00, -1.6418e+00, 1.7963e+00, -2.0671e+00]],
[[-2.4150e+00, 2.9655e+00, 2.6004e-01, -1.7344e+00, -4.8468e+00,
-1.6555e+00, -1.8036e+00, 1.5898e+01, -2.0078e+00, -1.9563e+00,
-5.5235e+00, -6.9839e+00, -2.2330e+00, -1.9891e+00, -2.4521e+00,
-2.3151e+00, 2.6808e+00, -4.2703e+00, -2.1547e+00, -4.0710e-01,
-1.7177e+00, -2.3013e+00, 2.7612e+00, -2.3445e+00, -1.9045e+00,
2.9985e+00, -1.7977e+00, -1.8816e-02, -1.6960e+00]],
[[-1.4256e+00, 1.0906e+00, 3.3821e+00, -1.5478e+00, -1.6942e+00,
-1.5854e+00, -2.4417e-01, 2.7178e+00, -1.2530e+00, -1.4479e+00,
-3.7336e+00, -7.0572e+00, -1.1972e+00, -1.6188e+00, -4.2215e+00,
-3.8705e+00, 1.6215e+01, -5.0594e-01, -1.2779e+00, 7.0080e-01,
-1.8873e+00, -1.7079e+00, -6.8577e+00, -1.5080e-03, -1.5215e+00,
1.2660e+00, -1.2671e+00, -1.3414e+00, -1.4342e+00]],
[[-8.9704e-01, 1.8598e+01, 3.8363e+00, -7.6451e-01, -2.0665e+00,
-7.1585e-01, 3.8746e+00, -7.9766e-01, -1.1897e+00, -3.4330e-01,
-7.6859e+00, -7.4154e+00, -7.5932e-01, -9.3892e-01, -1.5778e+00,
3.8570e+00, 2.7336e+00, 2.6286e+00, -3.8035e-01, -1.0038e+00,
-8.5127e-01, -1.0872e+00, -2.5204e+00, 9.3966e-01, -9.4037e-01,
-3.5173e+00, -5.6645e-01, -6.5331e+00, -9.3239e-01]]])
predict= tensor([[[25]],
[[17]],
[[15]],
[[ 7]],
[[16]],
[[ 1]]])
mans -> women
enc_input= tensor([[[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) dec_input= tensor([[[1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) n_hidden= 128
output.data= tensor([[[-2.8283, -2.3626, -0.1669, -2.7062, -4.9340, -2.6686, 1.1879,
-0.4988, -2.5126, -2.7524, -0.8902, -0.8576, -2.1348, -2.0980,
1.3107, -5.7952, 0.4938, -1.6269, -2.5288, 14.2410, -3.2220,
-2.8979, 0.5059, 0.8293, -2.9407, 0.1144, -2.8318, -2.2397,
-2.8485]],
[[-2.7481, 0.1070, -0.2417, -2.6742, -3.5322, -2.7356, -2.0069,
-0.2905, -2.8494, -2.4679, 0.4449, -0.2035, -2.8989, -2.5681,
-1.2044, 0.7797, -1.2203, 1.3424, -2.0246, 1.0619, -2.7756,
-2.6097, -0.8477, 14.0103, -2.4926, -3.1346, -3.0548, -2.6245,
-2.6824]],
[[-3.0481, 0.2810, -1.4473, -1.8614, -3.3435, -2.6207, -5.5615,
15.5710, -2.8886, -2.0847, -3.2127, 0.4409, -2.4811, -2.2347,
-3.1258, 2.3124, -3.0728, -3.4695, -2.5387, -2.7836, -2.3338,
-2.6220, 2.1170, 1.8315, -2.4378, 2.5099, -2.3928, -0.9716,
-2.1889]],
[[-2.3486, 2.1898, -0.0518, -1.6043, -4.2200, -1.6043, -3.1870,
16.6410, -2.0038, -1.8146, -4.4528, -5.3590, -1.9544, -1.7610,
-1.7343, -3.0047, 4.0441, -5.1262, -1.9987, -0.3312, -1.8511,
-2.2385, 1.3912, -2.0107, -1.8433, 2.5781, -1.7077, -0.8625,
-1.4298]],
[[-1.1249, 2.6187, 3.2088, -1.2680, -1.8667, -1.3058, 0.2252,
2.6478, -1.0426, -1.0876, -3.6653, -7.0211, -0.8896, -1.3647,
-3.9822, -3.6295, 15.3443, -0.4771, -0.9410, 0.4940, -1.5341,
-1.3913, -6.1356, 0.1833, -1.1524, 0.3265, -1.0332, -1.8967,
-1.1357]],
[[-0.6864, 18.7087, 3.2266, -0.6188, -2.0001, -0.5869, 3.7535,
-0.3294, -1.0357, -0.1939, -8.1245, -7.4850, -0.6122, -0.7862,
-1.3038, 4.1259, 2.6617, 2.5700, -0.1909, -1.1126, -0.6744,
-0.9144, -2.6198, 0.8000, -0.7998, -3.4453, -0.3744, -6.2346,
-0.7107]]])
predict= tensor([[[19]],
[[23]],
[[ 7]],
[[ 7]],
[[16]],
[[ 1]]])
king -> queen
enc_input= tensor([[[0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) dec_input= tensor([[[1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) n_hidden= 128
output.data= tensor([[[-2.0689, -8.7098, -4.8898, -1.6018, 2.4492, -1.4456, -2.5575,
1.2529, -1.5553, -1.4590, 0.8418, 1.3503, -1.8665, -1.5689,
-0.1531, -1.7443, -1.7615, -6.4119, -1.6575, -1.2020, -2.1042,
-1.6743, -1.8010, -0.4804, -1.3815, 15.5830, -1.6576, 0.2545,
-1.9206]],
[[-1.8489, -7.2973, 1.5389, -1.4012, -1.7340, -1.4996, -0.6075,
-4.5550, -2.0124, -1.3346, 15.1152, 1.9257, -1.5385, -1.3881,
0.6504, -5.3116, 0.2489, 2.4873, -1.6146, 0.9236, -1.7539,
-1.6810, 2.1537, 1.2465, -1.5198, -2.3739, -1.1541, -0.0671,
-1.3845]],
[[-1.7372, -9.3584, 2.6335, -1.4762, 2.5457, -1.0611, -4.0521,
0.1155, -1.6446, -1.7453, 1.8787, 15.4588, -1.1961, -1.3059,
0.2514, 3.1567, -8.6701, -0.8013, -1.5303, -0.6646, -1.6321,
-1.8999, -0.3239, 0.4056, -1.6625, 0.3260, -1.5085, 3.0456,
-1.6100]],
[[-1.2088, 1.6715, 0.6705, -0.9774, -7.1186, -1.0423, -0.8351,
3.2168, -1.2495, -1.0304, 1.5145, 0.5656, -1.1819, -1.0560,
-1.5316, -1.0357, -6.6219, -1.4356, -1.3063, 0.9956, -1.1131,
-1.3230, 15.0737, -0.2129, -1.1119, -0.9749, -1.0831, 0.7700,
-1.3064]],
[[-1.6763, -5.5338, 2.3379, -0.7959, -1.2500, -1.2617, -5.5603,
15.5783, -1.2844, -1.3196, -2.3954, 1.8382, -1.3597, -0.8014,
-3.4878, -1.6384, 3.2091, -6.1552, -1.2064, 0.5083, -0.8395,
-1.4700, -2.6530, 0.3554, -1.0272, 0.8245, -1.1982, 2.1557,
-0.9269]],
[[-0.7122, 17.1544, 3.1363, -0.6060, -3.1694, -0.4530, 2.9015,
1.3534, -0.7256, -0.1687, -4.6771, -6.5512, -0.6577, -0.5677,
-1.7814, 0.7493, 1.3163, -0.4522, -0.4733, -0.3166, -0.8819,
-0.8310, 0.9430, 1.2143, -0.3004, -2.3155, -0.3772, -7.6786,
-0.6709]]])
predict= tensor([[[25]],
[[10]],
[[11]],
[[22]],
[[ 7]],
[[ 1]]])
black -> white
enc_input= tensor([[[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) dec_input= tensor([[[1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]]) n_hidden= 128
output.data= tensor([[[-2.2227, 1.1039, 0.9361, -1.7916, 0.9939, -1.8225, 14.8655,
-6.5431, -2.2393, -2.3357, -1.2254, -2.7620, -2.2117, -2.2088,
2.6410, -2.9425, -0.4038, 0.8187, -1.6252, 1.9952, -1.7778,
-2.2960, -2.2505, -0.8751, -1.6060, 0.0786, -2.1974, -7.0257,
-2.2020]],
[[-1.6809, 2.2360, -1.8729, -1.2541, -2.6357, -1.6801, 0.3821,
-2.6765, -1.3480, -1.4661, -2.2543, -2.3800, -1.3111, -1.1596,
1.9145, 2.8206, 0.6113, 16.3379, -1.1935, -0.3985, -1.2032,
-1.4498, -2.1519, 1.9817, -1.3006, -5.3040, -1.1795, -4.1014,
-1.0693]],
[[-3.2444, -0.5409, -5.0188, -2.8395, -0.0612, -2.9714, -1.1149,
2.3132, -2.9405, -2.2777, -6.1815, -4.3277, -2.7849, -2.7514,
-0.9576, 2.2393, -0.0854, -2.6481, -2.7784, -1.7263, -3.5978,
-3.3209, -1.3265, -0.5880, -3.0756, 14.5987, -2.7313, -1.5224,
-3.0121]],
[[-2.4068, -0.7567, 2.4065, -2.1517, -3.5193, -2.1153, -1.1563,
2.7205, -2.1444, -2.0500, 0.1838, -7.6772, -2.0536, -2.5851,
-3.0046, -5.9824, 14.7826, 1.0193, -1.8719, 0.4687, -2.2422,
-2.4837, -2.6767, -0.7756, -2.2014, 1.1184, -2.0253, 0.7376,
-2.3218]],
[[-2.4552, 1.6956, 15.1841, -2.6467, 0.6999, -2.0372, 0.4377,
-1.0396, -2.6377, -2.2472, -4.1860, -0.9231, -2.2536, -2.4956,
-3.8611, 1.2366, 1.9323, -1.3519, -2.2468, -0.7968, -3.0316,
-2.3819, -5.4405, -1.0079, -2.5828, -1.3621, -2.2209, -2.9338,
-2.3236]],
[[-0.8805, 18.8412, 3.7769, -0.8079, -3.3419, -0.2910, 3.2971,
1.4113, -0.8959, -0.4190, -8.3294, -7.0438, -0.9324, -0.9681,
-1.5665, 3.6978, -0.3938, 3.2983, -0.7220, -0.8588, -0.9130,
-1.1669, -0.0783, 0.1662, -0.8183, -3.6403, -0.5167, -5.5895,
-0.7154]]])
predict= tensor([[[ 6]],
[[17]],
[[25]],
[[16]],
[[ 2]],
[[ 1]]])
up -> down
代码来自
https://github.com/wmathor/nlp-tutorial/blob/master/4-1.Seq2Seq/Seq2Seq_Torch.ipynb
这篇关于pycharm安装pytorch报错 提示系列问题 torch 包找不到因为pip版本低，结果升级了pip从19.3到20.2 4又提示高版不支持torch安装的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！
