https://www.52pojie.cn/thread-717061-1-1.html
Windows 10 startup item repair an operating system was’t found solution
Recently, in the dual systems of deepin15 ( Ubuntu ) and windows10 , we encountered a series of problems when uninstalling deepin. The records are as follows:
after windows installs deepin system, the general boot item will change to deepin boot. After uninstalling, the boot item will disappear, and the system will not automatically restore win boot, resulting in failure to enter the system normally. Before deleting the partition of deepin, you can use the official uninstall instruction of deepin to guide me to use it. However, it is invalid, so I have to delete the partition by myself, and then use the USB flash disk to start and repair it.
(step 1-2) it needs to be made on another computer that can be used normally
- Download laomaotao or u start or Chinese cabbage. And install. Insert the U-disk and make it with one click. When finished, unplug the U-disk. Insert the USB flash disk into the computer that can't be started normally, and press
F12 to select the option of USB flash disk. Enter (the name is usually the manufacturer of your USB flash disk). Select Windows 8 to start the new computer. Wait for normal entry. Open diskgenius, change the drive letter to the previous good state (generally starting from the U disk will change the initial drive letter), and use the boot repair software to automatically repair. Exit and restart. If the win system can be started normally after 6, the following can be ignored. Otherwise, if an operating system wasn't found, press Ctrl + Alt + Del to restart. Then restart or repeat the above 3-5 (no 6). And follow the steps below
- after entering the PE system, click the windows logo in the lower left corner and find dism + + in the folder. Open the top column, find the partition where the system is located, and click to find the tool repair guide. Restart, that's it.
Difference between vs code user and system version
Difference between vs code user and system version
Recently, the version of win10 LTSC has been re installed, and the new version of visual studio code has just come out.
However, the official website provides user and system versions. Simple installation is tried.
For installation
The
- user version can’t create the default folder in the system partition (disk)
Microsoft vs Code , which shows that the permissions are insufficient. The solution is to name the self created folder in the directory 
in advance, and then click next, For example, vs Code , in order to synchronize the path box, you need to delete the default Microsoft, as shown in figure 
For the system version, there is no such problem, but the next step is the same as other software
Language problems (win10 LTSC Chinese language)
I'm not sure whether this is just me or a common problem. For the user version of the installation program and vscode program, the default is English. It seems that the system language environment can't be read. The system version is the Chinese installation program directly, and the Chinese display panel user version uses the Chinese method
open vs code, enter Ctrl + Shift + P , and enter configure display in the search box Language , click open to change the quotation mark en to 'zh CN', or delete the quotation mark "en" to the colon, and enter Z to complete it as "locale": "zh CN" click the plug-in, click the bottom icon on the left side of the software, and then search Chinese , download the language pack, install and restart
Concluding remarks
at present, I don't know the specific difference between the two versions.
in stakoverflow's Q & A, some people said that there are differences between the two versions. I don't know for the moment. I may need to understand the good intentions of Microsoft. The software is also divided into users and system administrators, for ordinary users, in order to avoid unnecessary trouble, it is recommended to use the system version. For reference only
Solution of insufficient memory space or desktop stack in GX works2
Solution of insufficient memory space or desktop stack in GX works2
Open Mitsubishi PLC programming software GX works2 prompt
There is not enough memory space or desktop stack to start GX works2
Please restart GX works2 after finishing other applications
New on December 30, 2018, shared by @ Oops. You can delete the registry and reload it.
The method is as follows
- press
win + R , enter regedit , and press enter to find HKEY_ CURRENT_ Right click to delete. Reload the software. Solutions that cannot be opened
Google and Baidu have found many similar situations, but there is no perfect solution
For example, reloading software still doesn't work. It's also said that the cost of reloading the system is too high. After all, a lot of software has to be installed.
#Solution:
##1. First install GX works3
?2. Then install GX works2. Perfectly solve this problem
?3. Pay attention not to unload GX3, otherwise GX2 still can't work.
I thought the GX3 version was high and good, but later I learned that 2 and 3 only support different PLC. The FX2N and gxworks2 that I wanted to use were available, so I unloaded 3. As a result, 2 is not available. Later in accordance with the previous way to try, and good..
Hope to help you
Vs (Visual Studio) encountered a solution that could not open iostream
Recently, when configuring some projects, I couldn’t even open the iostream header file
Finally, we try to reset vs C + +, and successfully solve the problem.
My version is vs2013 in the following way
Tools -> Import and Export -> Reset all -> ..
Tools → import and export settings → reset all settings → next →
Select “save the current settings (I chose to reset directly)”,
Next → select “development language to reset (for example, Visual C + + development settings)” → finish
My CSDN first blog actually wrote this..
Inheritance relationship causes class loading order
Question: when wechat authentication defines variables in weixinpaycontroller, the assigned value is obtained by reading the configuration file in initcontroller, which may cause that sometimes the data of the configuration file cannot be read?
Reason: This is caused by the loading order of the controller. First, the class inheritance relationship of weixinpaycontroller is weixinpaycontroller — & gt; absauthcontroller — & gt; basemcccontroller — & gt; mvccontroller. The inheritance relationship of initcontroller class is initcontroller — & gt; mvccontroller. It can be seen that these two controllers are inherited from mvccontroller, so there will be the problem of loading order first and then
Short term solution: when deploying online, test it first, and you can see that there is no problem. If there is no problem, you can. If there is a problem, it will be deployed again, and then detected again
Long term solution: weixinpaycontroller is directly inherited from initcontroller, so that the loading order is fixed and consistent
How to prevent duplicate submission
When the user clicks the submit button twice in a row, if no measures are taken, there will be two pieces of data in the data.
Solution: check when inserting database. Insert a distributed lock on the database in the code. Use redis as the object of the lock. After locking, make a judgment. The short rule is to look up a piece of data from the database. If the data does not exist, insert it and generate an ID in the database. If it exists (insert the second data), take out this data and update it. In this way, the problem of repeated insertion can be solved.
Existing problems: it is not clear whether it is a duplicate submission or a user’s update operation
[Solved] volatile was removed and now has no effect. Use `with torch.no_grad():` instead.
Leave a reply
Solution: volatile was removed and now has no effect. Use with torch.no_grad():instead.
Source code
self.priors = Variable(self.priorbox.forward(), volatile=True)
the reason
It volatilehas been removed in the torch version .
Before pytorch 0.4.0 input = Variable(input, volatile=True) set volatile to True, as long as an input is volatile, the output is also volatile, which can guarantee that there is no intermediate state; but canceled after pytorch 0.4.0 The volatile mechanism is replaced with functions such as torch.no_grad(), torch.set_grad_enable(grad_mode)
torch.no_grad() is a context manager.
When using pytorch , not all operations require the generation of calculation graphs (the construction of the calculation process to facilitate gradient back propagation and other operations). For the calculation operation of tensor, the default is to construct the calculation graph. In this case, you can use with torch.no_grad(): to force the subsequent content not to construct the calculation graph.
The torch.no_grad() will affect pytorch’s backpropagation mechanism. In the test, because it is determined that backpropagation will not be used, this mode can help save memory space. The same is true for torch.set_grad_enable(grad_mode)
change into
with torch.no_grad():
self.priors = Variable(self.priorbox.forward())
or
self.priors = Variable(self.priorbox.forward())How to Solve Error: Unkown Mixed bake mode in LightModeUtil.MapSettings() UnityEditor.DockArea:OnGUI()
Goto the Setting as follow:
Just remove the check
If this method does not works for you. please let me know.
How to Solve TypeError: type numpy.ndarray doesn‘t define __round__ method
Pycharm Warning TypeError: type numpy.ndarray doesn’t define round method
1.the problem
matrix = round(np.array([[1.5,2.3],[1,2],[1,2]]))
The error interface
type is wrong: the type numpy.ndarray does not define the __round__ method
2.solve
Change the round method to np.round
code show as below:
matrix = np.round(np.array([[1.5,2.3],[1,2],[1,2]]))
How to Solve TypeError: Fetch argument None has invalid type
problem:
Solution:
It may be because sess.run([op])the op operation in does not exist, that is, it is None.
[Solved] Jupyter notebook: TypeError: __init__() got an unexpected keyword argument ‘io_loop’
When opening a Python file today, an error message was reported:
TypeError: __init__() got an unexpected keyword argument’io_loop’
Obviously it is a file copied from the old computer to the new computer. The previous operation was OK. Why did it report an error when running on the new computer?
wrong reason:
When configuring the python environment, the default tornado version is the latest version (it happens that my new computer reconfigured the python environment, so I installed the latest version), but the io_loop parameter was discarded after version 4.0.
solution:
1. First uninstall the currently installed tornado
pip uninstall tornado
2. Install the lower version of tornado
pip install tornado==4.1
After the installation is successful, re-execute the python file and it can run normally