Comments (8)
Then, if it's okay to ship numpy.h
, I will prepare a PR to fix that and also to add bool pinned
to ReadTensor
.
from dali.
Hi @fversaci,
Please add REBUILD_BUILDERS=YES
to the build.sh
invocation to rebuild cmake with https support (NVIDIA/DALI_deps#99) needed to pull nvImageCodec.
from dali.
Hi, I've created the PR. Please have a look and let me know if it looks okay to you or if any changes are needed.
from dali.
Just merged. It should be available in the next nightly build.
from dali.
Hi @fversaci,
Thank you for reaching out.
- The
numpy.h
header file does not seem to be available to the user (it's not being installed here), so we included it in our code asdali_numpy.h
. Would it be possible to make it available for the users as well?
@szkarpinski, @klecki do you remember what was the reason for not shipping this file?
- The original
ReadTensor
function always creates tensors in pinned memory. We want to be able to use it also in normal memory, so we extendedReadTensor
toNewReadTensor
, which also accepts abool pinned
argument. If that doesn't break anything you're using, it might be useful to have it directly in the originalReadTensor
function.
It could be just an oversight. As I see this function is used only in tests. Feel free to post a PR that would fix this.
- We use
NewReadTensor
in the decoder in a serial way. Is this function thread-safe, and could we use it in parallel as well?
As far as I see the functions used are stateless and the only side effect I see is the change of the InputStream state. So if you want to share the same InputStream between multiple concurrent threads it won't work.
from dali.
@JanuszL it looks like an oversight for me with regards to the numpy.h, but @azrael417 or @mzient may know more about it.
from dali.
Hi,
I'm trying to compile DALI (using docker builder) to prepare the PR, but i get an error.
Am doing something wrong?
I'm launching this
git clone --recursive https://github.com/NVIDIA/DALI
cd DALI/docker
CUDA_VERSION=12.2 CREATE_RUNNER=yes ./build.sh
which causes this
CMake Error at nvimgcodec_headers-subbuild/nvimgcodec_headers-populate-prefix/src/nvimgcodec_headers-populate-stamp/download-nvimgcodec_headers-populate.cmake:1
59 (message):
Each download failed!
error: downloading 'https://developer.download.nvidia.com/compute/nvimgcodec/redist/nvimgcodec/linux-x86_64/nvimgcodec-linux-x86_64-0.2.0.6-archive.tar.xz'
failed
status_code: 1
status_string: "Unsupported protocol"
log:
--- LOG BEGIN ---
Protocol "https" not supported or disabled in libcurl
Closing connection -1
--- LOG END ---
error: downloading 'https://developer.download.nvidia.com/compute/nvimgcodec/redist/nvimgcodec/linux-x86_64/nvimgcodec-linux-x86_64-0.2.0.6-archive.tar
.xz' failed
status_code: 1
status_string: "Unsupported protocol"
log:
--- LOG BEGIN ---
Protocol "https" not supported or disabled in libcurl
Closing connection -1
[...]
gmake[2]: *** [nvimgcodec_headers-populate-prefix/src/nvimgcodec_headers-populate-stamp/nvimgcodec_headers-populate-download] Error 1
gmake[1]: *** [CMakeFiles/nvimgcodec_headers-populate.dir/all] Error 2
gmake: *** [all] Error 2
CMake Error at /usr/local/share/cmake-3.18/Modules/FetchContent.cmake:987 (message):
Build step for nvimgcodec_headers failed: 2
Call Stack (most recent call first):
/usr/local/share/cmake-3.18/Modules/FetchContent.cmake:1082:EVAL:2 (__FetchContent_directPopulate)
/usr/local/share/cmake-3.18/Modules/FetchContent.cmake:1082 (cmake_language)
cmake/Dependencies.common.cmake:293 (FetchContent_Populate)
cmake/Dependencies.cmake:108 (include)
CMakeLists.txt:293 (include)
-- Configuring incomplete, errors occurred!
See also "/opt/dali/build-docker-Release-122_x86_64/CMakeFiles/CMakeOutput.log".
See also "/opt/dali/build-docker-Release-122_x86_64/CMakeFiles/CMakeError.log".
from dali.
Perfect, thanks!
from dali.
Related Issues (20)
- Segmentation fault when using 'mixed' HOT 5
- Bbox Pruning Too Aggressive? HOT 5
- Indexing video with binary mask HOT 1
- source_info tensor not guaranteed to contain correct data HOT 1
- 16 bit gray scale Image read error HOT 1
- COCO Reader pixelwise_masks Emtpy Output HOT 7
- Dali on Jetson: nvidia.dali.fn.readers.video_resize is missing HOT 4
- Numpy reader test (GDS) HOT 4
- How to add a scalar value to the loader? HOT 1
- Can DALI be integrated into HuggingFace Trainer? HOT 9
- Bug in creating `TensorGPU` when `stream` key is `None` in CUDA array interface HOT 2
- Configure max image size HOT 3
- Webdataset reader behavior with many sources HOT 1
- ModuleNotFoundError: No module named 'nvidia.dali.python_function_plugin' HOT 3
- Speed up Dino with DALI HOT 3
- error using webdataset
- webdataset cannot stop cycling at end of epoch HOT 11
- Get audio data from external data sources and start iteration HOT 7
- Encountered some issues when using mirror flip HOT 2
- Can AArch64 Numba tests be re-renabled? HOT 2
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 dali.