Latest version of numpy causing issues, constraining it to v1.* #35
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mariodruiz
reviewed
Jun 17, 2024
setup.py
Outdated
| @@ -41,7 +41,7 @@ | |||
| packages=find_packages(), | |||
| python_requires="==3.9.*", | |||
| install_requires=[ | |||
| "numpy", | |||
| "numpy==1.*", | |||
There was a problem hiding this comment.
I think it is more readable if numpy<2.0
mariodruiz
approved these changes
Jun 19, 2024
STFleming
added a commit
to STFleming/Riallto
that referenced
this pull request
Jun 26, 2024
* Fix link to Ryzen AI webpage (AMDResearch#31) * extra symbols to wsl path resolution (#10) * Latest version of numpy causing issues, constraining it to v1.* (AMDResearch#35) * Latest version of numpy causing issues, constraining it to v1.* * readability fix * Enable Kernel fusion (AMDResearch#34) * Include library of kernels in the compilation * Fix typos * Include kernels header file * Prepare kernels for superkernel * Include test name * Initial superkernel test * Simplyfy test * Add second superkernel * Rename test to better reflect the nature of PR * Rename test to better reflect the nature of PR * Align code * Fix call * More fixes * Remove fake kernels * Kernel fusion working!!!! * Fix duplicate variable name * Add another test * Remove unnecesary flag * Remove * revert to original * Revert to original * Revert to original * Revert to original * Include kernels * Copy kernel * Import shutil * Allow to load the same Riallto app up to 4 times (AMDResearch#36) * Allow multiple instances of the same app to be run * Try to fix test * Add more tests * Flake8 * More appropiate test name * Make sure AppRunner has device, device is not found is previous call had an space issue * Made code more pythonic * Handle AppRunner object even if it fails * Flake8 * Report app start column (AMDResearch#37) * Allow multiple instances of the same app to be run * Try to fix test * Add more tests * Flake8 * More appropiate test name * Make sure AppRunner has device, device is not found is previous call had an space issue * Made code more pythonic * Handle AppRunner object even if it fails * Flake8 * Try to add start column * Add starting column * Flake8 * Update check * Fix where assert is done * Fix issue with counting * Check that xbutil count is working * Do not run xbutil test for Linux * Add missing os import * Skip test in Linux --------- Co-authored-by: Sarunas Kalade <[email protected]> Co-authored-by: Shane Fleming <[email protected]>
STFleming
added a commit
to STFleming/Riallto
that referenced
this pull request
Jun 27, 2024
* Added pyxrt linux bindings from @skalade Removing windows python bindings Remove check from linux for now Dropping down to python3.8 for mlir-aie compatibility in the same env. Still having issues not being able to find the CU name in the xclbin when loading the application commented out state check as it is not working with linux driver (different pyxrt bindings needed?) Call the script to run unregistered xclbins before in the apprunner init Copy headerfiles one by one, on linux subprocess isn't liking the wildcards. Now checks to see if it is a windows filepath rather that just running on windows when doing path translations. Added a simple test to test various path types. Getting tests up and running on linux Changes to arguments to get the latest NPU linux driver working Rebuilding all the prebuilt applications with the patched version of MLIR-AIE for linux Stopping nputop erroring out for now Attempting to bump linux support to py3.10 bumping pyxrt to 310 bindings (thanks Sarunas) Falling back to older driver version for RyzenAI SW Linux tests Constraining the appropriate python version based on platform. Signing does not need to happen on windows based flows, so adding a check. Also, keeping the error status check after an apprun for windows but this does not currently work with the bindings on linux. Reintroducing the windows python bindings alongside the linux one Added docker setup for Riallto Linux Keep stability check in for Windows Case matters for linux which caught this issue. * Add citation file (#7) * Add citation file * Shane's orcid * Adds an aditional check to see if the license file is on a floating license server. (#9) * kernelbuilder falls over if the cache file does not exist * Fix for color threshold v1 * temporarily disabling onnx test to try out CI flow * Trying a self-hosted linux runner. * Removing linux actions for now * Adding more of the install steps into the fork, such as the driver builder. * Added install steps * Adding opendownloads link for python 3.10 tarballs * Migrating to py312 for Ubuntu24.04 support * Removing the need to register xclbins, not required in later xdna-driver versions * Docker files and scripts bumped to Ubuntu 24.04 * first cut at fully fleshing out the build script that will provision an Ubuntu 24.04 system from scratch * Added scripts for launching jupyter lab and pytests * Fixing issues with the LIC Arg in the dockerfile * Fix issues encountered on first bringup on wiped machine * Changed the name of the setup script * Updated the readme to reflect 24.04 changes * F/B from Shawn * Time estimate on Riallto docker build * Some clarifications * Fix installed packages was at the wrong point * More explicit copying of header files into temporary build area * Keep things cleaner in a temporary directory while building the docker image * Correct docker engine setup link * Temporary files included in Dockerfile now * Missing temp folder * workaround for fix-missing issue * Add user to the video group for webcam access from within a container. * Adjusting permissions and changing the launch_jupyterlab script to use Riallto/notebooks instead * Forgot perl! * Revert "Add user to the video group for webcam access from within a container." This reverts commit 0055155. * add extra exceptions when doing device check using gc.get_objects() * keep only 1 exception in the loop, I don't remember why I added the first one * Now using 6.10_rc2 of the Linux kernel * Install steps README update for 6.10 * numpy 2.0 release is causing issues so adding constraint here * update wheel/binary raw links to v1.0 branch commit amd/RyzenAI-SW@c9d3db1 * update ipu driver image * update driver url * add script to update ryzenai-sw packages * fix upgrade script path and add error checking * add check to look if ryzenai-sw zip already extracted * change python installation to all users * Updating to the latest installer.zip file * grabbing from the correct directory * Permissions on video files within the container * Linuxwebcam fix (#5) * Set video backend depending on OS * Remove unused code * Add wait to avoid race condition * Use platform.system() * Testing out a linux CI script * selecting the right runner * Getting the right directory structure * Setting correct flag for windows CI runner * Removing tty * Add bfloat16 support (#6) * add bfloat16 support * add ml_dtypes to requirements * Correct filtering of platform for nputop * Eula agreement check was dropped so adding that back in * Updating the README to point to the Linux install steps * enter venv in script (#9) * This needs to fail, the kill at the top of the script should be sufficient * Applying some of Marios feedback to this markdown also * Prepare for release, merge main (#7) * Fix link to Ryzen AI webpage (AMDResearch#31) * extra symbols to wsl path resolution (#10) * Latest version of numpy causing issues, constraining it to v1.* (AMDResearch#35) * Latest version of numpy causing issues, constraining it to v1.* * readability fix * Enable Kernel fusion (AMDResearch#34) * Include library of kernels in the compilation * Fix typos * Include kernels header file * Prepare kernels for superkernel * Include test name * Initial superkernel test * Simplyfy test * Add second superkernel * Rename test to better reflect the nature of PR * Rename test to better reflect the nature of PR * Align code * Fix call * More fixes * Remove fake kernels * Kernel fusion working!!!! * Fix duplicate variable name * Add another test * Remove unnecesary flag * Remove * revert to original * Revert to original * Revert to original * Revert to original * Include kernels * Copy kernel * Import shutil * Allow to load the same Riallto app up to 4 times (AMDResearch#36) * Allow multiple instances of the same app to be run * Try to fix test * Add more tests * Flake8 * More appropiate test name * Make sure AppRunner has device, device is not found is previous call had an space issue * Made code more pythonic * Handle AppRunner object even if it fails * Flake8 * Report app start column (AMDResearch#37) * Allow multiple instances of the same app to be run * Try to fix test * Add more tests * Flake8 * More appropiate test name * Make sure AppRunner has device, device is not found is previous call had an space issue * Made code more pythonic * Handle AppRunner object even if it fails * Flake8 * Try to add start column * Add starting column * Flake8 * Update check * Fix where assert is done * Fix issue with counting * Check that xbutil count is working * Do not run xbutil test for Linux * Add missing os import * Skip test in Linux --------- Co-authored-by: Sarunas Kalade <[email protected]> Co-authored-by: Shane Fleming <[email protected]> * Added defensive check for key URLs at the start of script to prevent mid-way failure * Added docker url to checklist after Joshua issue * Stop it from blocking jupyter lab from starting if no webcam is connected * Updating the launch jupyter script so that the notebooks directory is an argument. Also added waits after the docker container stop commands to prevent race conditions when attempting to start the next container instance. * add extra check to confirm wsl image downloaded (#11) * Notebook update to factor in Linux support (#8) * Notebook update to factor in Linux support * Addressing some of Mario's feedback * Added the cell outputs * Pushing correct notebook where outputs should be left in * Added command to check the kernel version on Linux * Stop it from blocking jupyter lab from starting if no webcam is connected * Removing the heading from alert boxes * fix mlirsequencebuilder dtype assignment for egress (#12) :fire: * Fix visualisation for data parallel apps (#13) * Add test to check data parallel with non tested buffer names * Fix memory tile animations for buffers that have name different than in_buffer * Do not track svg files * Update year * Move function to the top * Add more automatic tests * Attempt to fix buffer representation * Fix checks * Fix token generation * Intermediate update * Fix visualization for passthrough mode * Improve quality of tests * Add randomness to the buffers name to make sure nothing is hardcode * Add more files to ignore only in current folder * Add buffer when data comes from non neighbor * Fix asserts * Add checks * Start in a different place --------- Co-authored-by: Mario Ruiz <[email protected]> Co-authored-by: skalade <[email protected]> Co-authored-by: skalade <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Describe the problem solved by the commit
Numpy 2.0 is causing some issues and the current setup.py does not constrain the version.
How is the problem solved?
Constrain setup.py to use numpy==1.*
Are there any risks associated to the change?
None
Is there a documentation impact?
None
Checklist
Please provide screenshots (if applicable)
Related issues