Comments (10)
Hi etfong,
Thanks for your feedback! I guess that your TensorBoard log files are not on drive C.
When creating a TensorBoard process, we pass the full path of the log directory to TensorBoard. However, TensorBoard uses a colon as a separator between the optional "run name" and the path in the --logdir argument. So if your log directory is e.g. "d:\log", then TensorBoard will interpret this string as two parts due to the ":". The run name is "d", and the log directory is "\log". Since TensorBoard is launched from VS, the drive letter of the default working directory is "C" and then TensorBoard cannot find log files.
As a workaround, please put your projects and log files on drive C. We will fix the issue soon.
from vs-tools-for-ai.
Fixed in version 0.3.3.0.
Thanks.
from vs-tools-for-ai.
I have the same problem.(http://msi:1025)
And put the projects and log files on drive C does not work.
AI Tools 0.4.1.664470
Visual Studio 2017 15.7.5
from vs-tools-for-ai.
@XUEZIJIAN , thank you very much for the feedback.
Did you see a TensorBoard process is launched? If not, please open a CMD terminal, go to the log directory, and run "tensorboard --logdir=." to check if errors happen.
Which version of TensorFlow are you using? I suggest that you install TensorFlow 1.5.0 on Windows and try it again.
Thanks.
from vs-tools-for-ai.
@linmajia ,
Sorry to reply late.
I use tensorflow1.6. Actually it works fine two month ago(by right click in Visual Studio).
The process is just flashed. I can't see the content. After about 10s, the browser will open automatically.(msi:1025/)
And I don't know how to use cmd in VS. But I tried several:
PS D:> python -m tensorboard --logdir=logs/
C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\python.exe: No module named tensorboard.main; 'tensorboard' is a package and cannot be directly executed
PS D:> tensorboard --logdir=logs/
Fatal error in launcher: Unable to create process using '""c:\program files (x86)\microsoft visual studio\shared\anaconda3_64\python.exe" "C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\Scripts\tensorboard.exe" --logdir=logs/'
PS D:> python -m tensorflow.tensorboard --logdir=logs/
C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\lib\site-packages\h5py_init_.py:34: FutureWarning: Conversion of the second argument of issubdtype from float
to np.floating
is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type
.
from ._conv import register_converters as _register_converters
C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\python.exe: No module named tensorflow.tensorboard
Thank you very much.
from vs-tools-for-ai.
@XUEZIJIAN , thanks for the runtime information.
It seems that you are using Anaconda shipped with VS 2017. Did you install other Python environments too? First, you need to confirm that TensorFlow 1.6 and TensorBoard 1.6 are really installed within this Anaconda environment. Then, you open a CMD terminal and run: "C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\Scripts\tensorboard.exe" --logdir=logs/ "
from vs-tools-for-ai.
@linmajia
Thank you very much for your patience.
They're in the same environment. but the tensorboard is 1.10 now.
here's the result
C:\Users\XZJ> "C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\Scripts\tensorboard.exe" --logdir=D:/logs
Fatal error in launcher: Unable to create process using '""c:\program files (x86)\microsoft visual studio\shared\anaconda3_64\python.exe" "C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\Scripts\tensorboard.exe" --logdir=D:/logs'
It looks like still I didn't use CMD correctly? Vs can get a address at least.
from vs-tools-for-ai.
@XUEZIJIAN , did you install TensorFlow/TensorBoard directly into the Anaconda master environment? Seems that TensorFlow/TensorBoard are broken. I suggest that you uninstall and then reinstall them:
- Open a CMD terminal with Administrative permission.
- "C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\python.exe" -m pip uninstall -y -q tensorboard tensorflow
- "C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\python.exe" -m pip install tensorflow==1.6
If you have NVidia GPU, you can replace "tensorflow" with "tensorflow-gpu" to install the GPU version.
from vs-tools-for-ai.
@linmajia ,yes and I just reinstalled them according to your guidance.
Now I have tensorflow-gpu1.6 and tensorboard1.6.
But still have the problem.
I wonder why it can run and get the address by Visual Studio tool, and why my port is not 6006.
Thank you.
from vs-tools-for-ai.
@XUEZIJIAN , Tools for AI just launches TensorBoard using a random available port to avoid port confliction. The issue is that your TensorBoard in Anaconda is broken.
My suggestion is that you install Python from www.python.org, set it as the default Python environment in VS and install TensorFlow. You can refer here
from vs-tools-for-ai.
Related Issues (20)
- Error installing due to prereq failing HOT 3
- mnist.py Bug
- AI Tools Freezing VS HOT 2
- AI VS Tool convert giving onnx v4 instead of v7 HOT 1
- Cannot find .tar.gz package when submitting Python job HOT 2
- VS crashes when opening remote Jobs view
- Can not run with it HOT 3
- cannot solve the model.dll
- Please support CUDA 10 HOT 3
- Mac OS support
- new ModelManager Error HOT 8
- Unexpected inference results
- model.RunModel hangs when model is 'Optimized'
- Inference project wizard shows incorrect warning
- Support for Visual Studio 2019 HOT 6
- The AI-Tools option was not found in the Visual Studio menu bar
- VS w/ Turkish Language Pack shows Tools for AI Start Page in Chinese HOT 2
- 合作开发关于chat GPT 内核的应用 HOT 2
- This repo is missing a LICENSE file
- This repo is missing important files HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from vs-tools-for-ai.