oucmachinelearning/oucml

oucml's Issues

PDF中很多图片没有

if the code search table can support command +F "ctrl + F", it will be better.

those code search table is picture, not easy for searching. the code search table would support command +F "ctrl + F" will be better.

赞赞赞

哈哈哈，写的非常好啊！我 FORK 啦~~~

超链接错误

Dive-into-DL-PyTorch.pdf

pdf中出现了大量的超链接错误
比如：
超出范围
多个连接只有一个有效

Undefined name 'DictAsMember' in config.py

flake8 testing of https://github.com/OUCMachineLearning/OUCML on Python 3.7.1

$ flake8 . --count --select=E9,F63,F72,F82 --show-source --statistics

./GAN/人脸还原SCFEGAN/utils/config.py:23:21: F821 undefined name 'DictAsMember'
            value = DictAsMember(value)
                    ^
1     F821 undefined name 'DictAsMember'
1

E901,E999,F821,F822,F823 are the "showstopper" flake8 issues that can halt the runtime with a SyntaxError, NameError, etc. These 5 are different from most other flake8 issues which are merely "style violations" -- useful for readability but they do not effect runtime safety.

• F821: undefined name name
• F822: undefined name name in __all__
• F823: local variable name referenced before assignment
• E901: SyntaxError or IndentationError
• E999: SyntaxError -- failed to compile a file into an Abstract Syntax Tree

3.2.1 RuntimeError: expected scalar type Double but found Float

It reports "expected scalar type Double but found Float" with torch.mm in chapter 3.2.1.
Fixed by add "to (torch.float32)" as bellow:
features = torch.from_numpy(np.random.normal(0, 1, (num_examples, num_inputs))).to(torch.float32)

关于测试

您好，关于SRGAN方面
是否有测试板块

10.5.2节（P313）出现大量数学公式报错

R.T.

3.3.4 TypeError: 'LinearNet' object is not subscriptable

Hi,
thanks for the hard work!

I am trying this part but it gives me error:

---

TypeError Traceback (most recent call last)
in
----> 1 init.normal_(net[0].weight, mean=0, std=0.01)
2 init.constant_(net[0].bias, val=0)

TypeError: 'LinearNet' object is not subscriptable

this is the linearnet:

class LinearNet(nn.Module):
def init(self, n_feature):
super(LinearNet, self).init()
self.linear = nn.Linear(n_feature, 1)

def forward(self, x):
    y = self.linear(x)
    return y

net = LinearNet(num_inputs)
print(net)

my Torch version is 1.5.1

您好，我最近也在学习srgan，使用的是eriklindernoren/srgan的脚本。

他的博客里只写了train的脚本，请问您有test的脚本么 或者您是怎么做前向传播的？