Please make sure that this is a bug. As per our
GitHub Policy,
we only address code/doc bugs, performance issues, feature requests and
build/installation issues on GitHub. tag:bug_template

System information

• Have I written custom code (as opposed to using a stock
  example script provided in TensorFlow): Yes

cc @ezyang @anjali411 @dylanbespalko

What's the ETA for updating the massively outdated documentation?

Please update all documents that are related building CNTK from source with latest CUDA dependencies that are indicated in CNTK.Common.props and CNTK.Cpp.props.
I tried to build from source, but it's a futile effort.

Here is some general documentation improvement suggestion regarding flatbuffers. This is what I remember struggling with (from the top of my head). Also some point might be a little bit opinionated.

Some of the points were taken from #5387.

• Regarding the builder:
  • The fact that its kinda like an arena allocator used for serialization.
  • The concept of offsets, what do they repre

It would be beneficial to write a design document on INT8 DNNL implementation and put it into
FluidDoc next to other Design documents:
https://github.com/PaddlePaddle/FluidDoc/tree/develop/doc/fluid/design/mkldnn

One unit test in the R package is currently broken. Steps to reproduce on Mac

export CXX=/usr/local/bin/g++-8 CC=/usr/local/bin/gcc-8
Rscript build_r.R
cd R-package/tests
Rscript testthat.R

This results in the following error at the ends of the logs

[LightGBM] [Info] Saving data to binary file /var/folders/xq/wktq4zdx4jd3qdpk34d28m940000gn/T//RtmpiY1DzV/lgb.Dataset_1555

Bug report

Describe the bug

Here is a clear and concise description of what the problem is:

Hello, There is a bug on Kodi (tested on Windows Store version)
Using Alt+Space I can move Kodi on full-screen mode

Expected Behavior

Here is a clear and concise description of what was expected to happen:

Actual Behavior

Possible Fix

To Reproduce

Steps to reproduce t

The new converter (#1151) uses member objects to operate on data from the .dat files. While these provide the necessary functions for conversion, they are also significantly larger than the original dat file entries (20x - 400x the size). This results in a comparatively high memory usage, especially for DE2 which uses ~1.6 GB of memory for all dat file entries. Note that the converter does not

Any plan to add the solarized color theme? Or there are any documents to show how to adding color theme?

Is there a default seed? I copied the Python code from seed_test.py and removed the set_seed calls. The test passed thousands of times despite no set_seed calls being made. Is this expected? If so, can we document this? Is there an explicit default seed?

Just as a sanity check, I modified the calls to set the seed to two distinct values and it failed every time, as expected.

The reason I'm a

According to Assimp's documentation (see anim.h, line 269). The description of aiNodeAnim says the following:

The transformation matrix computed from these values replaces the node's original transformation matrix at a specific time. This means all keys are absolute and not relative to the bone default pose.

From my understanding, I believe the original transformation refers to aiNode's

This class could be used instead of cd file https://catboost.ai/docs/concepts/input-data_column-descfile.html when creating Pool from filez. The class should have init function, methods load and save, and Pool init method should be able to use object of this class instead of cd file during initialization.

Now we follow the official documents to build and test tensorflow serving with latest dev docker image. The examples can not be run successfully because the tensorflow version is 2.1.0 in tensorflow/serving:latest-devel while the examples code like mnist_saved_model.py still uses TensorFlow 1.x APIs.

We should upgrade the examples to TensorFlow 2 so that users can build and test with latest

I'm working with Open3D with data from different devices. One of them does not provide the point cloud as a file, so I am creating the file by myself with a little Python script.

However, I am confused by the supported formats:
The pts format contains lines of [x, y, z, i, r, g, b]. xyz and rgb are intuitive and explained, but what exactly is the i for? And in what format is it exp

One of the instructions for compiling from source https://github.com/mapnik/mapnik/wiki/UbuntuInstallation is to install Harfbuzz (http://www.freedesktop.org/software/harfbuzz/release/harfbuzz-0.9.34.tar.bz2), in my version of Ubuntu 19.10 this caused a problem with libpango, in turn causing gdm to break and stopped my machine from booting.

It might be worth attaching a note to the wiki th

So, we have some existing code over at the PyTorch Ignite project that is actually pretty general and might be really handy to have in DALI: pytorch/ignite#766

In core PyTorch, you can chain transformations and FileIO together easily with the Compose() operation: https://pytorch.org/docs/stable/torchvision/transforms.html#torchvision.transforms.Compose

Something like

I tried running the .ipynb associated with the Entropy Maximization example at cvxpy website. The notebook uses CVXOPT as the solver. When I run the same notebook, I am getting the error that ends with the message:

SolverError: Either candidate conic solvers (['CVXOPT']) do not suppo

This is a feature request (also open to helping contribute) to add touch input to the pywin32 API.
It appears there is an equivalent input method to mouse_event, InjectTouchInput which would allow pywin32 to emulate touch. T

The following code fails:

from panda3d.core import *

dst = PNMImage(2, 2)
src = PNMImage(1, 1)
dst.add_sub_image(src, 1, 1, 0, 0, 1, 1)

However, using copy_sub_image works fine. This is because add_sub_image does not properly add the offset when sampling the source image.

Fixing this will require adding the offset calculation from the other *_sub_image methods to

As code on master is Python3 only depndency on six should be removed and affected code needs to be adapted to use Python3 constructs.

The current portable build script is getting messy and unnecessarily complex. We should separate the Linux build of portable out of the python script and into our Scons build scripts. The portable python build script then would be dedicated to windows.

To Reproduce

See the portable build script.

Expected behavior

Portable for Windows is built with the dedicate python script a

I am having quite a bit of trouble trying to create a minimum working example for the RTS game, in which I can get a single state, visualize its variables and pass an action back to the environment.

I took a look at the Basic Usage section of the README.md, but it looks like the code is not supposed to be used, just interpreted as pseudocode. For example, I don't know where functions such as

Cocos2d is a framework for building 2D games, demos, and other graphical/interactive applications:

https://github.com/los-cocos/cocos